Ovii
A grainy vintage computer with a terminal on its screen

Developer Survey

2026 edition

How developers work changes year to year. Explore the tools they rely on, how AI fits into daily practice, and the paths their careers take through the 2026 survey responses.

Relationship to coding

Are you someone who creates code? Please select one of the following options that best describes you today.

All respondents · n = 30,794
  • Creating code is my primary job
    75.5%23,234
  • I create code occasionally
    15.4%4,743
  • I manage or support coders
    3.8%1,170
  • I do not create code
    3.1%944
  • I review code, rarely write it
    2.3%703

Educational attainment

Which of the following best describes the highest level of formal education that you've completed?

All respondents · n = 30,761

  • Bachelor's degree
    i
    B.A., B.S., B.Eng., etc.
    40.3%12,396
  • Master's degree
    i
    M.A., M.S., M.Eng., MBA, etc.
    26.0%7,990
  • Some college/university study without earning a degree
    12.4%3,813
  • Secondary school
    i
    e.g. American high school, German Realschule or Gymnasium, etc.
    7.9%2,431
  • Professional degree
    i
    JD, MD, Ph.D, Ed.D, etc.
    5.9%1,812
  • Associate degree
    i
    A.A., A.S., etc.
    3.1%942
  • Primary/elementary school
    2.3%706
  • Other, please specify
    2.2%671

Employment status

Which of the following best describes your current employment status?

All respondents · n = 30,903

  • Employed
    65.6%20,259
  • Independent contractor, freelancer, or self-employed
    14.9%4,607
  • Student
    10.2%3,139
  • Not employed
    5.6%1,742
  • Retired
    2.0%607
  • I prefer not to say
    1.8%549

Employment additional status

Which of the following other current work, school, or life responsibilities are you currently engaged in?

All respondents · n = 28,099

  • None of the above
  • Caring for dependents (children, elderly, etc.)
  • Engaged in paid work (20-29 hours per week)
  • Attending school (full-time)
  • Volunteering (regularly)
  • Engaged in paid work (less than 10 hours per week)
  • Attending school (part-time)
  • Engaged in paid work (10-19 hours per week)
  • Transitioning to retirement (gradually reducing work hours)
  • Employed21,594
  • Student4,059
  • Independent contractor, freelancer, or self-employed5,655
  • Not employed1,883
  • Retired707
View exact data
Exact respondent counts for each flow
From / toNone of the aboveCaring for dependents (children, elderly, etc.)Engaged in paid work (20-29 hours per week)Attending school (full-time)Volunteering (regularly)Engaged in paid work (less than 10 hours per week)Attending school (part-time)Engaged in paid work (10-19 hours per week)Transitioning to retirement (gradually reducing work hours)
Employed8,7633,9543,4416461,9941,031916609240
Student200643082,27436728131122925
Independent contractor, freelancer, or self-employed1,3668761,085232641468256407324
Not employed939209671142271151115744
Retired26075201119034172080

Country

Where do you live?

All respondents · n = 21,370

  • United States of America
    22.3%4,755
  • Germany
    9.3%1,983
  • United Kingdom of Great Britain and Northern Ireland
    6.4%1,374
  • India
    6.1%1,309
  • Canada
    4.0%851
  • France
    3.7%794
  • Netherlands
    2.7%577
  • Poland
    2.5%527
  • Australia
    2.4%522
  • Ukraine
    2.3%484

Work Setting

Which best describes your current work situation?

All respondents · n = 24,602

  • Remote
    34.8%8,564
  • Hybrid (some remote, leans heavy to in-person)
    19.9%4,883
  • In-person
    17.6%4,335
  • Hybrid (some in-person, leans heavy to flexibility)
    15.5%3,823
  • Your choice (very flexible, you can come in when you want or just as needed)
    12.2%2,997

Industry

What industry is the company you work for in?

All respondents · n = 18,909

  • Software development
    20.0%3,785
  • IT services, consulting, systems integration, or managed services
    15.4%2,914
  • Other, please specify
    7.1%1,344
  • Banking/financial services
    6.0%1,126
  • Healthcare
    5.1%955
  • Retail, e-commerce, or consumer services
    5.0%952
  • Fintech
    4.5%853
  • Education
    4.2%796
Show all 23 resultsShow fewer results
  • Engineering services or applied R&D
    4.0%749
  • Manufacturing
    4.0%749
  • Government
    3.9%734
  • Scientific research
    3.0%570
  • Media or advertising services
    2.9%542
  • Utilities, internet, or telecommunications
    2.9%542
  • Transportation or supply chain
    2.6%485
  • Energy
    2.2%410
  • Insurance
    1.6%294
  • Hospitality or travel
    1.2%226
  • Non-profit or social services
    1.1%209
  • Agriculture, forestry, fishing, and hunting
    1.1%207
  • Life sciences or biotechnology
    0.9%173
  • Mining or construction
    0.8%153
  • Broadcasting or publishing
    0.8%141

1.2 Role functions

IC or PM

Are you an individual contributor or people manager?

All respondents · n = 24,749

  • Individual contributor
    50.8%12,568
  • Individual contributor in a senior or staff-level role
    30.4%7,517
  • Both people manager and individual contributor
    15.5%3,828
  • People manager
    3.4%836

Years coding

Including any education, how many years have you been coding in total? Please round to the nearest whole number, excluding any decimal points.

All respondents · n = 25,204

  • 0-4 years
    9.8%2,467
  • 5-9 years
    19.9%5,021
  • 10-14 years
    19.1%4,821
  • 15-19 years
    13.6%3,432
  • 20-29 years
    20.3%5,128
  • 30-39 years
    10.0%2,512
  • 40+ years
    7.2%1,823

Years of work experience

How many years have you been working professionally in total? Please round to the nearest whole number, excluding any decimal points. If you have no professional work experience, you can leave this question unanswered.

All respondents · n = 24,607

  • 0-4 years
    22.8%5,612
  • 5-9 years
    18.4%4,519
  • 10-14 years
    17.3%4,268
  • 15-19 years
    12.3%3,030
  • 20-29 years
    17.5%4,296
  • 30-39 years
    7.8%1,929
  • 40+ years
    3.9%953

Title

Which of the following best describes your current role, or the role you held for the most time in the past year?

All respondents · n = 25,767

  • Developer, full-stack
    27.3%7,042
  • Developer, back-end
    13.5%3,473
  • Student
    7.1%1,823
  • Architect, software or solutions
    7.0%1,792
  • Developer, desktop or enterprise applications
    4.0%1,030
  • Developer, front-end
    3.6%933
  • Developer, embedded applications or devices
    2.8%731
  • Academic researcher
    2.7%705
Show all 41 resultsShow fewer results
  • Developer, mobile
    2.7%685
  • Engineering manager
    2.3%603
  • Other
    2.1%545
  • DevOps engineer or professional
    2.0%527
  • AI/ML engineer
    1.8%455
  • Data engineer
    1.7%426
  • Senior executive (C-suite, VP, etc.)
    1.5%374
  • Founder or entrepreneur
    1.4%350
  • Retired
    1.3%330
  • Developer, game or graphics
    1.2%316
  • Data scientist
    1.2%301
  • System administrator
    1.2%299
  • Data or business analyst
    1.0%254
  • Solutions engineer, implementation engineer, or forward-deployed engineer
    0.9%244
  • Cybersecurity or InfoSec professional
    0.9%228
  • Software engineer
    0.8%219
  • Cloud infrastructure engineer
    0.8%218
  • Developer, QA or test
    0.8%203
  • Developer, AI apps or physical AI
    0.8%200
  • Applied scientist
    0.7%183
  • Educator or instructor
    0.7%175
  • Project manager
    0.6%158
  • Product manager
    0.6%149
  • Hardware engineer
    0.6%145
  • Support engineer or analyst
    0.5%123
  • Site reliability engineer
    0.5%119
  • UX, research, or UI design professional
    0.4%97
  • Database administrator or engineer
    0.4%94
  • Developer advocate or developer relations professional
    0.2%61
  • Financial analyst or engineer
    0.2%45
  • Technical writer or documentation professional
    0.2%41
  • Platform engineer
    0.1%37
  • Marketing/Sales professional
    0.1%34

Functions at work

In a typical week, which of the following activities are part of your work? Select all that apply.

All respondents · n = 24,714

  • Writing, editing, or generating code
    89.9%22,221
  • Debugging or troubleshooting
    83.2%20,573
  • Reviewing code, pull requests, or technical changes
    66.7%16,494
  • Making technical design or architecture decisions
    63.4%15,679
  • Answering questions from teammates, customers, users, or stakeholders
    59.7%14,754
  • Planning, scoping, or estimating work
    57.5%14,202
  • Writing or updating documentation
    56.3%13,918
  • Deploying or releasing software
    51.8%12,800
Show all 18 resultsShow fewer results
  • Writing or maintaining tests
    51.1%12,618
  • Changing production code, systems, or infrastructure
    49.9%12,334
  • Investigating an incident, outage, or production issue
    46.5%11,486
  • Monitoring systems, logs, metrics, or alerts
    44.3%10,955
  • Maintaining dependencies, packages, or development environments
    39.3%9,720
  • Coordinating work across teams or functions
    32.3%7,974
  • Finding or fixing security, privacy, or compliance issues
    31.4%7,754
  • Onboarding to a new project, tool, system, or codebase
    23.2%5,732
  • Analyzing data, reports, or business metrics
    23.2%5,722
  • None of the above
    3.4%834

1.3 Employer + technology details

Employer organization size

Approximately how many people are employed by your employer? This should only include your primary company. If you had more than one employer, please consider the one you spent most of your time working for in the past year.

All respondents · n = 21,115

  • 10,000 or more employees
    13.9%2,941
  • 5,000 to 9,999 employees
    4.8%1,017
  • 1,000 to 4,999 employees
    11.4%2,409
  • 500 to 999 employees
    6.2%1,307
  • 100 to 499 employees
    16.5%3,474
  • 20 to 99 employees
    17.4%3,677
  • Less than 20 employees
    16.8%3,555
  • Just me - I am a freelancer, sole proprietor, etc.
    10.5%2,213
  • I don’t know
    2.5%522

Tech budget status

Has your department's budget increased, decreased, or stayed the same in your most recent budget cycle? When considering your department, refer to the one you are working in today.

All respondents · n = 20,915

  • I don't know
    32.3%6,757
  • Stayed the same
    27.9%5,839
  • Increased
    22.5%4,713
  • Decreased
    17.2%3,606

Relationship to purchase decisions at work

Which of the following describes your role in choosing or using software tools at your organization? Select all that apply.

All respondents · n = 20,042

  • I influence tool choices by giving feedback or requirements
    59.5%11,916
  • I evaluate, recommend, or shortlist tools
    48.2%9,664
  • I use tools selected by others
    45.6%9,131
  • I make the final decision to purchase tools
    17.7%3,552
  • I do not have a role in choosing or using software tools at my organization
    16.6%3,320
  • I approve or manage budget for tools
    9.3%1,867

Reasons to endorse a technology

Thinking about technologies you enjoy using for work, school, or personal projects, what is the main reason you would recommend one to others? Select one.

All respondents · n = 22,371

  • It is reliable, performant, and stable
    17.3%3,874
  • It helps me be more productive or save time
    15.9%3,563
  • It has a strong community, ecosystem, or open-source presence
    12.9%2,888
  • It has a strong developer experience, such as APIs, SDKs, tooling, or integrations
    11.4%2,541
  • It is flexible, customizable, or easy to adapt to my needs
    7.5%1,678
  • It has a reputation for quality and long-term trustworthiness
    7.5%1,666
  • It has excellent documentation, examples, or learning resources
    6.1%1,370
  • It is easy to learn and use
    5.5%1,226
Show all 14 resultsShow fewer results
  • I’m not sure
    5.5%1,223
  • Other, please specify
    2.8%632
  • It has useful AI features or agent capabilities
    2.3%505
  • It has strong security, privacy, or compliance features
    1.9%423
  • It has manageable cost, pricing, or licensing
    1.9%421
  • I would not recommend any technology I use
    1.6%361

Count of tools used

In a typical month over the past year, about how many distinct software tools, platforms, or services did you personally use to do your primary work? Count tools you intentionally opened, configured, or relied on to create, analyze, manage, build, deploy, support, or share work. These may be free, purchased, internal, or open-source tools. Do not count general operating systems, web browsers, programming languages, libraries/packages, one-off scripts or commands, or multiple interchangeable utilities that serve the same narrow purpose.

All respondents · n = 14,836

  • 5
    17.2%2,550
  • 10
    16.1%2,389
  • 3
    9.6%1,427
  • 4
    8.5%1,263
  • 6
    7.7%1,135
  • 8
    6.1%910
  • 7
    5.2%775
  • 20
    4.8%708
Show all 25 resultsShow fewer results
  • 2
    4.6%687
  • 15
    4.4%646
  • 12
    3.4%501
  • 1
    1.9%275
  • 9
    1.8%265
  • 30
    1.3%191
  • 0
    1.0%153
  • 25
    0.9%135
  • 11
    0.8%114
  • 50
    0.6%89
  • 13
    0.6%85
  • 14
    0.5%77
  • 16
    0.4%58
  • 40
    0.3%42
  • 100
    0.3%41
  • 17
    0.2%35
  • 18
    0.2%35

1.4 Job satisfaction + AI

Job satisfaction

How satisfied are you in your current professional role?

All respondents · n = 18,876

  • Complacent at Work
    45.1%8,516
  • Not Happy at Work
    32.6%6,161
  • Happy at Work
    22.2%4,199

Reasons to be "just okay" with current job

If you feel complacent at work, what is the primary driver?

All respondents · n = 8,312

  • I do not feel complacent
    34.8%2,889
  • Burnout/tech fatigue
    21.1%1,753
  • Economic uncertainty making me hesitant to switch roles
    18.6%1,545
  • AI automating the challenging parts of my job
    9.9%822
  • Lack of upward mobility due to corporate tightening
    9.8%811
  • Other
    5.9%492

Reasons your job is easy

In the past 6 months, which factor has done the most to make your job easier to do?

All respondents · n = 18,521

  • Nothing has made my job easier in the past 6 months
    21.1%3,902
  • Access to the right tools or platforms
    20.1%3,714
  • Supportive teammates, managers, or mentors
    17.2%3,191
  • Clear priorities or requirements
    7.1%1,314
  • Better processes, workflows, or coordination
    6.8%1,256
  • I’m not sure
    6.0%1,107
  • Enough time, staffing, or resources
    5.9%1,088
  • Other
    5.6%1,044
  • Helpful documentation or knowledge sharing
    4.1%768
  • Reliable systems, infrastructure, or environments
    4.0%741
  • Leadership activity or organizational changes
    2.1%396

Does AI make work easier?

For each factor that made your job easier, was AI or AI agents a reason?

Additional question

In the past 6 months, which factor has done the most to make your job easier to do?

All respondents · n = 11,973

  • Yes, AI or AI agents were the main reason
  • No, AI or AI agents were not a reason
  • Yes, AI or AI agents contributed, but were not the main reason
  • I’m not sure
  • Access to the right tools or platforms3,688
  • Supportive teammates, managers, or mentors3,164
  • Clear priorities or requirements1,307
  • Enough time, staffing, or resources1,078
  • Better processes, workflows, or coordination1,242
  • Reliable systems, infrastructure, or environments735
  • Helpful documentation or knowledge sharing759
View exact data
Exact respondent counts for each flow
From / toYes, AI or AI agents were the main reasonNo, AI or AI agents were not a reasonYes, AI or AI agents contributed, but were not the main reasonI’m not sure
Access to the right tools or platforms2,49321396814
Supportive teammates, managers, or mentors1472,11384757
Clear priorities or requirements7384036925
Enough time, staffing, or resources11755837429
Better processes, workflows, or coordination24247550817
Reliable systems, infrastructure, or environments12434925012
Helpful documentation or knowledge sharing11931431313

Reasons your job is difficult

In the past 6 months, which factor has done the most to make your job harder to do?

All respondents · n = 18,214

  • Unclear priorities or requirements
    16.6%3,029
  • Inefficient processes, workflows, or coordination
    14.6%2,658
  • Not enough time, staffing, or resources
    14.1%2,566
  • Leadership activity or organizational changes
    11.9%2,162
  • Nothing has made my job harder in the past 6 months
    9.6%1,743
  • Unreliable systems, infrastructure, or environments
    8.4%1,523
  • Poor documentation or knowledge sharing
    7.1%1,292
  • Other
    6.1%1,108
  • I’m not sure
    4.7%865
  • Lack of support from teammates, managers, or mentors
    4.2%761
  • Lack of access to the right tools or platforms
    2.8%507

Does AI make your job more difficult?

For each factor that made your job harder, was AI or AI agents a reason?

Additional question

In the past 6 months, which factor has done the most to make your job harder to do?

All respondents · n = 12,246

  • No, AI or AI agents were not a reason
  • Yes, AI or AI agents contributed, but were not the main reason
  • Yes, AI or AI agents were the main reason
  • I’m not sure
  • Unclear priorities or requirements3,009
  • Inefficient processes, workflows, or coordination2,638
  • Not enough time, staffing, or resources2,554
  • Poor documentation or knowledge sharing1,283
  • Unreliable systems, infrastructure, or environments1,514
  • Lack of support from teammates, managers, or mentors749
  • Lack of access to the right tools or platforms499
View exact data
Exact respondent counts for each flow
From / toNo, AI or AI agents were not a reasonYes, AI or AI agents contributed, but were not the main reasonYes, AI or AI agents were the main reasonI’m not sure
Unclear priorities or requirements2,084597202126
Inefficient processes, workflows, or coordination1,81256020561
Not enough time, staffing, or resources1,75351119298
Poor documentation or knowledge sharing86922912362
Unreliable systems, infrastructure, or environments83234425385
Lack of support from teammates, managers, or mentors5171436227
Lack of access to the right tools or platforms2821099018

1.5 Compensation

Compensation by role, and top responding countries

What is your current total annual compensation (salary, bonuses, and perks, before taxes and deductions)? Please enter a whole number in the box below, without any punctuation. If you are paid hourly, please estimate an equivalent yearly salary. If you prefer not to answer, please leave the box empty.

All respondents · n = 12,040

  • Engineering manager
    $133,263378 respondents
  • Senior executive (C-suite, VP, etc.)
    $131,498176 respondents
  • Cloud infrastructure engineer
    $110,000129 respondents
  • Architect, software or solutions
    $102,3551,023 respondents
  • Solutions engineer, implementation engineer, or forward-deployed engineer
    $101,710144 respondents
  • Software engineer
    $101,218105 respondents
  • AI/ML engineer
    $95,337194 respondents
  • Founder or entrepreneur
    $94,012111 respondents
Show all 22 resultsShow fewer results
  • Data scientist
    $90,982155 respondents
  • Data engineer
    $90,000233 respondents
  • Developer, back-end
    $85,2961,922 respondents
  • Developer, embedded applications or devices
    $85,296412 respondents
  • DevOps engineer or professional
    $85,096307 respondents
  • Developer, mobile
    $84,277346 respondents
  • Developer, desktop or enterprise applications
    $80,000559 respondents
  • Developer, full-stack
    $79,6103,761 respondents
  • Developer, game or graphics
    $77,002108 respondents
  • Developer, front-end
    $70,320441 respondents
  • Data or business analyst
    $67,161119 respondents
  • Academic researcher
    $61,413253 respondents
  • System administrator
    $55,230122 respondents
  • Other
    $51,000106 respondents

· All respondents: limit bar charts to roles with 100 or more responses; Country-level: limit bar charts to roles with 30 or more responses. All reflect median salary after being converted to USD.

Compensation by experience

What is your current total annual compensation (salary, bonuses, and perks, before taxes and deductions)? Please enter a whole number in the box below, without any punctuation. If you are paid hourly, please estimate an equivalent yearly salary. If you prefer not to answer, please leave the box empty.

Additional question

How many years have you been working professionally in total? Please round to the nearest whole number, excluding any decimal points. If you have no professional work experience, you can leave this question unanswered.

All respondents · n = 12,040

By developer role
Responses1053,761
$40k$60k$80k$100k$120k$140k121416182022Engineering managerSenior executi…Cloud infrastructure engin…Architect, software…Solutions engineer,…Software engineerAI/ML engineerFounder or entrepreneurData scientistData engineerDeveloper, back-endDeveloper, embedded applic…DevOps enginee…Developer, mobileDeveloper, desktop or ente…Developer, full-stackGame/graphicsDeveloper, front-endData or business analystAcademic researcherSystem administratorOtherMedian yearly salary (USD)Average years of professional experience
View exact data
Exact median salary and experience values
GroupYears of experienceMedian salary (USD)Respondents
Engineering manager18.5$133,263378
Senior executive (C-suite, VP, etc.)21.9$131,498176
Cloud infrastructure engineer16.2$110,000129
Architect, software or solutions20.1$102,3551,023
Solutions engineer, implementation engineer, or forward-deployed engineer17.2$101,710144
Software engineer15.2$101,218105
AI/ML engineer12.1$95,337194
Founder or entrepreneur18.5$94,012111
Data scientist13$90,982155
Data engineer15.4$90,000233
Developer, back-end14.5$85,2961,922
Developer, embedded applications or devices14.5$85,296412
DevOps engineer or professional14.2$85,096307
Developer, mobile14.2$84,277346
Developer, desktop or enterprise applications16.7$80,000559
Developer, full-stack13.9$79,6103,761
Developer, game or graphics12.4$77,002108
Developer, front-end11.8$70,320441
Data or business analyst14.6$67,161119
Academic researcher12.6$61,413253
System administrator14.5$55,230122
Other15.7$51,000106

Only groups with at least 100 responses are shown. Median salaries are converted to USD.

Compensation by experience

What is your current total annual compensation (salary, bonuses, and perks, before taxes and deductions)? Please enter a whole number in the box below, without any punctuation. If you are paid hourly, please estimate an equivalent yearly salary. If you prefer not to answer, please leave the box empty.

Additional question

How many years have you been working professionally in total? Please round to the nearest whole number, excluding any decimal points. If you have no professional work experience, you can leave this question unanswered.

All respondents · n = 12,040

By country
Responses1072,885
$0$20k$40k$60k$80k$100k$120k$140k$160k8101214161820United States of AmericaSwitzerlandAustraliaDenmarkNorwayUnited Kingdom of G…CanadaGermanyAustriaNetherlandsNew ZealandBelgiumSwedenSpainPolandFrancePortugalCzech RepublicItalyBrazilUkraineIndiaMedian yearly salary (USD)Average years of professional experience
View exact data
Exact median salary and experience values
GroupYears of experienceMedian salary (USD)Respondents
United States of America17.4$154,0002,885
Switzerland14.7$139,262229
Australia17.9$104,550306
Denmark14.2$104,214123
Norway15.5$103,004110
United Kingdom of Great Britain and Northern Ireland17.8$96,333855
Canada16.7$94,315539
Germany14.2$85,2961,218
Austria13.4$84,912163
Netherlands15.1$82,453360
New Zealand19.9$80,858107
Belgium13.8$74,964117
Sweden16.6$71,965241
Spain16.1$68,237331
Poland12.7$65,733309
France13.5$62,550526
Portugal12.7$61,413135
Czech Republic13$56,438158
Italy15.7$47,766302
Brazil15.5$36,002244
Ukraine10.2$35,800259
India8.8$18,802472

Only groups with at least 100 responses are shown. Median salaries are converted to USD.

AI

2.1 AI usage

AI tool usage

Do you currently use AI tools or AI agents in your role at work?

All Respondents · n = 17,464

  • AI coding assistants or coding agents
    65.9%11,509
  • General-purpose AI chat tools
    62.5%10,920
  • AI agents or automated workflows
    26.2%4,582
  • Internal AI tools built by my company
    17.8%3,115
  • I don’t use AI tools
    17.2%3,003
Reference chart

AI tool usage frequency

How often do you use each of these AI tools or AI agents in your role at work?

All Respondents · n = 14,478

  • Daily
  • Weekly
  • Monthly or Infrequent
  • AI coding assistants or coding agents
  • General-purpose AI chat tools
  • AI agents or automated workflows
  • Internal AI tools built by my company
View exact data
Percent and respondent count for All Respondents
ActivityDailyWeeklyMonthly or Infrequent
AI coding assistants or coding agents73.0% (8,403)20.1% (2,316)6.2% (712)
General-purpose AI chat tools65.6% (7,162)26.8% (2,922)7.0% (768)
AI agents or automated workflows63.1% (2,889)25.5% (1,168)10.3% (470)
Internal AI tools built by my company47.7% (1,487)30.2% (941)20.7% (645)

Overall usage: 77% daily, 18% weekly, and 5% monthly or infrequent.

Reference chart

AI tool usage daily

You indicated that you use AI tools and/or agents daily. How much of your workday do you use AI tools to help you work?

All Respondents · n = 10,926

  • 4+ hours/day
    30.9%3,378
  • 2-4 hours/day
    24.1%2,636
  • 1-2 hours/day
    25.3%2,765
  • <1 hour every day
    12.2%1,335
  • <1 hour and not every day
    7.0%763
  • None
    0.5%49
Reference chart

Attitudes towards AI tools

How would you describe your attitude toward using AI tools or agents as part of your work?

All Respondents · n = 14,428

  • Very favorable
    31.6%4,564
  • Somewhat favorable
    30.4%4,385
  • Neutral
    15.3%2,205
  • Somewhat unfavorable
    14.6%2,110
  • Very unfavorable
    6.9%994
  • Not sure
    1.2%170
Reference chart

2.2 Trust in AI and other trends

Trust in AI tools

How much do you trust the output from AI tools or AI agents as part of your workflow?

All Respondents · n = 14,304

  • I trust it when I can easily verify the output
    48.0%6,872
  • I trust it for many tasks, but not important work decisions
    16.3%2,329
  • I trust it for low-risk work tasks only
    16.3%2,326
  • I do not trust it for most work tasks
    12.3%1,754
  • I trust it for many tasks, including important work decisions
    6.6%949
  • Not sure / I do not use AI tools
    0.5%74
Reference chart

AI workflow current and expected

Which parts of your workflow are you currently integrating into AI or using AI agents to accomplish? Will they be more or less integrated over the next year? Please select one for each scenario.

All Respondents · n = 12,547

  • Current use
  • Expect more
  • Expect less
  • No plans to use
  • Not in workflow
  • Data, logs and metrics
  • Complex questions
  • Straightforward questions
  • Architecture and design
  • Dev and test environments
  • Debugging and refactoring
  • Deploying and operating
  • Security issues
Show all 19 resultsShow fewer results
  • Code in a familiar area
  • Code in an unfamiliar area
  • Visual and audio assets
  • Product or UX design
  • Code and design review
  • Internal knowledge
  • Onboarding to a codebase
  • Documentation
  • Emails and messages
  • Writing tests
  • Tickets and plans
View exact data
Percent and respondent count for All Respondents
ActivityCurrently use AI or AI agents for thisExpect to use AI More or AI agents more for thisExpect to use AI or AI agents less for thisDo not use AI or AI agents for this and do not plan toThis is not part of my workflow
Analyzing data, logs, metrics, or query results41.9% (4,848)37.5% (4,338)5.7% (664)16.3% (1,887)13.0% (1,502)
Answering complex or ambiguous technical questions50.0% (5,874)32.7% (3,843)11.4% (1,335)17.9% (2,098)4.4% (514)
Answering straightforward technical questions59.4% (7,017)29.0% (3,422)7.5% (881)16.5% (1,952)4.2% (500)
Architecture or technical design exploration47.4% (5,656)28.2% (3,358)8.2% (983)19.2% (2,292)12.1% (1,445)
Creating or configuring development/test environments28.1% (3,219)28.8% (3,298)6.7% (770)28.7% (3,282)17.9% (2,046)
Debugging, troubleshooting, or refactoring code63.8% (7,684)36.1% (4,340)8.2% (987)8.8% (1,065)2.7% (323)
Deploying, operating, or troubleshooting production systems19.9% (2,264)23.3% (2,651)7.4% (843)36.0% (4,093)22.0% (2,507)
Finding or fixing security issues34.6% (4,018)37.9% (4,409)6.6% (765)14.7% (1,711)19.1% (2,223)
Generating code in a familiar area69.3% (8,402)31.2% (3,784)7.7% (937)9.0% (1,091)2.3% (276)
Generating code in an unfamiliar area56.0% (6,707)33.8% (4,051)11.0% (1,320)12.3% (1,470)4.3% (511)
Generating visual, audio, or multimedia elements17.5% (1,997)18.8% (2,142)5.3% (608)24.1% (2,752)42.9% (4,896)
Product, UX, or visual design30.3% (3,551)25.4% (2,975)5.0% (583)15.1% (1,768)34.6% (4,050)
Reviewing code, pull requests, or design decisions40.5% (4,757)32.4% (3,811)8.7% (1,018)22.6% (2,657)10.1% (1,189)
Searching, summarizing, or interpreting internal knowledge49.7% (5,786)36.1% (4,199)5.7% (669)15.8% (1,844)7.9% (922)
Understanding or onboarding to a new codebase49.7% (5,870)35.1% (4,139)5.7% (678)12.9% (1,520)11.6% (1,373)
Writing documentation, comments, or knowledge base content51.1% (6,036)35.9% (4,239)7.0% (821)16.3% (1,928)6.7% (793)
Writing emails, messages, summaries, or other communication23.5% (2,700)20.8% (2,393)8.2% (940)45.9% (5,279)10.9% (1,255)
Writing or improving tests58.1% (6,915)35.0% (4,166)5.6% (661)8.5% (1,015)10.4% (1,239)
Writing tickets, requirements, plans, or project updates27.8% (3,187)28.0% (3,212)7.5% (862)30.5% (3,500)17.2% (1,973)
Reference chart

Next steps after receiving an AI prompt response

When you receive an AI-generated answer for a development task, what do you usually do before using it? Select all that apply.

All Respondents · n = 13,162

  • Run it locally
    76.5%10,067
  • Compare against codebase context
    63.9%8,411
  • Inspect tests/security implications
    53.0%6,977
  • Read official docs
    38.0%4,998
  • Search Stack Overflow
    12.2%1,601
  • Use as-is
    9.6%1,265
  • Ask a coworker
    6.9%907
  • I do not use AI tools.
    1.7%228
Reference chart

How good are AI tools at complex tasks

How well do the AI tools or AI agents you use in your development workflow handle complex tasks?

All Respondents · n = 14,341

  • Very well at handling complex tasks
    16.0%2,288
  • Good, but not great at handling complex tasks
    38.8%5,564
  • Neither good or bad at handling complex tasks
    14.4%2,067
  • Bad at handling complex tasks
    16.0%2,299
  • Very poor at handling complex tasks
    8.0%1,141
  • I don't use AI tools for complex tasks / I don't know
    6.9%982
Reference chart

AI tool count

In a typical week, about how many distinct AI tools or AI-enabled products do you personally use for work? Count a tool if you intentionally use its AI capability. This may include chatbots, AI search or answer tools, AI coding assistants, AI features in IDEs, autonomous AI agents, internal AI assistants, local model runtimes, or AI tools built into workplace software. Do not count AI features you do not knowingly use. Do not count multiple models inside the same product unless you use them through separate tools or interfaces.

All Respondents · n = 11,919

  • 0 tools
    0.6%75
  • 1 tool
    19.7%2,351
  • 2 tools
    34.4%4,097
  • 3-5 tools
    40.5%4,823
  • 6-10 tools
    3.8%450
  • 11+ tools
    1.0%123
Reference chart

2.3 AI agent tools

AI agent data storage tools

Have you used any of the following tools for AI agent memory or vector databases in the past year? Do you want to use any of the following in the next 6 months?

Used · n = 1,392

  • Mem0
    20.0%279
  • LangMem
    15.2%212
  • Memary
    12.6%176
  • Graphiti
    11.6%162
  • Supermemory
    11.6%162
  • Zep
    10.8%151
  • Letta
    10.8%150
  • Cognee
    9.6%133
  • Hyperspell
    9.3%129
Reference chart

AI agent orchestration tools

Have you used any of the following tools for AI agent orchestration or agent frameworks in the past year? Do you want to use any of the following in the next 6 months?

Used · n = 4,400

  • LangChain
    30.2%1,329
  • OpenAI Agents SDK
    23.8%1,048
  • Terminal Use
    18.9%832
  • LangGraph
    18.8%825
  • Llama Stack
    17.4%767
  • Google ADK
    14.0%617
  • Hermes Agent
    10.2%449
  • Vercel AI SDK
    9.0%395
Show all 22 resultsShow fewer results
  • Pydantic AI
    8.9%390
  • CrewAI
    5.5%242
  • Semantic Kernel
    5.0%218
  • AutoGen
    4.5%200
  • Strands
    4.0%178
  • DSPy
    3.6%159
  • Mastra
    3.0%132
  • Dify
    2.9%126
  • Smolagents
    2.6%116
  • Atomic Agents
    2.6%115
  • Instructor
    2.4%106
  • Agno
    2.4%105
  • 21st.dev
    2.3%100
  • GripTape
    2.1%91
Reference chart

AI agent observability and security

Have you used any of the following tools for AI agent observability, prompts, or evals in the past year? Do you want to use any of the following in the next 6 months?

Used · n = 2,277

  • Sentry
    26.0%592
  • LangSmith
    17.3%393
  • MLflow
    15.2%347
  • Langfuse
    15.2%345
  • Datadog LLM
    11.9%271
  • Weights & Biases
    10.7%244
  • Arize AI
    5.8%133
  • Braintrust
    5.8%132
Show all 32 resultsShow fewer results
  • DeepEval
    5.8%131
  • Respan
    5.6%127
  • Promptfoo
    5.3%121
  • Agenta
    5.0%114
  • Ragas
    4.9%111
  • Galileo AI
    4.8%110
  • Portkey
    4.8%109
  • LangWatch
    4.4%101
  • Helicone
    4.2%95
  • Humanloop
    4.1%93
  • PromptLayer
    4.0%92
  • Patronus AI
    4.0%91
  • HoneyHive
    4.0%90
  • Confident AI
    3.8%86
  • Traceloop
    3.8%86
  • Chamber
    3.7%85
  • Opik
    3.7%84
  • Future AGI
    3.5%80
  • Sentrial
    3.5%79
  • Athina AI
    3.4%78
  • Lunary
    3.4%77
  • Moda
    3.3%76
  • Ashr
    3.1%70
  • Parea AI
    3.0%69
Reference chart

AI coding agents

Have you used any of the following coding agents or assistants in the past year? Do you want to use any of the following in the next 6 months?

Used · n = 12,255

  • Claude Code
    65.5%8,030
  • GitHub Copilot
    58.7%7,196
  • OpenAI Codex
    29.5%3,620
  • Cursor
    25.5%3,127
  • Google Antigravity
    16.0%1,956
  • Gemini Code Assist
    14.6%1,789
  • Open Code
    13.6%1,662
  • JetBrains AI
    12.8%1,569
Show all 33 resultsShow fewer results
  • Windsurf
    6.5%796
  • Warp
    5.2%641
  • Cline
    4.5%551
  • Pi
    4.4%535
  • Amazon Q Developer
    3.9%476
  • Continue
    3.8%466
  • Mistral Vibe
    3.2%393
  • Kilo Code
    3.2%392
  • Devin
    2.9%360
  • Google Jules
    2.5%307
  • Replit Agent
    2.3%285
  • Tabnine
    2.1%257
  • Aider
    2.1%256
  • Codium (Qodo)
    1.7%206
  • BLACKBOX AI
    1.6%202
  • Supermaven
    1.3%158
  • Trae
    1.3%154
  • Augment Code
    1.2%145
  • Sourcegraph Cody
    0.9%114
  • Emdash
    0.7%85
  • Approxima
    0.6%69
  • ECA
    0.6%68
  • Sweep
    0.6%68
  • Kavia AI
    0.5%63
  • Syntropy
    0.5%63
Reference chart

AI-no-code agents

Have you used any of the following no-code agent builder tools in the past year? Do you want to use any of the following in the next 6 months?

Used · n = 2,219

  • Lovable
    38.3%850
  • Replit
    28.3%629
  • v0
    27.5%611
  • Bolt.new
    17.2%381
  • Base44
    11.3%250
  • Langflow
    7.8%174
  • Stack AI
    7.2%159
  • Flowise
    4.1%92
Show all 18 resultsShow fewer results
  • Voiceflow
    4.1%92
  • Create
    4.1%91
  • Glue
    3.8%85
  • Rivet
    3.5%78
  • Botpress
    3.5%77
  • Relevance AI
    3.2%72
  • Sparkles
    3.1%68
  • Wordware
    3.0%67
  • Marblism
    2.9%65
  • Pickaxe
    2.7%61
Reference chart

AI agent automation tools

Have you used any of the following workflow automation tools in the past year? Do you want to use any of the following in the next 6 months?

Used · n = 4,173

  • Make
    38.9%1,625
  • n8n
    32.7%1,364
  • Zapier
    18.5%771
  • Temporal
    6.0%249
  • Prefect
    3.0%125
  • Trigger.dev
    2.4%100
  • Inngest
    2.3%96
  • Gumloop
    2.2%93
Show all 13 resultsShow fewer results
  • Relay.app
    2.2%92
  • DBOS
    2.1%89
  • Activepieces
    2.0%82
  • Bubble Lab
    1.9%79
  • Cofia
    1.7%69
Reference chart

2.4 AI at work

AI impact in different work functions

When using AI tools or AI agents at work, how useful has AI proven itself for the task or activity? Select one answer for each.

All Respondents · n = 12,734

  • Not applicable
  • No strong opinion
  • More impractical than useful, not useful
  • Useful
  • Very useful
  • Architecture exploration
  • Communication
  • Debugging/refactoring
  • Design
  • Documentation
  • Implementation
  • Review
View exact data
Percent and respondent count for All Respondents
ActivityNot applicableNo strong opinionMore impractical than useful, not usefulUsefulVery useful
Architecture exploration12.6% (1,563)18.9% (2,348)9.4% (1,171)34.4% (4,274)24.7% (3,072)
Communication18.8% (2,337)23.6% (2,931)22.6% (2,804)21.8% (2,705)13.2% (1,637)
Debugging/refactoring4.8% (604)13.7% (1,720)8.8% (1,101)38.1% (4,776)34.5% (4,327)
Design17.2% (2,151)19.3% (2,423)14.3% (1,796)30.7% (3,844)18.5% (2,322)
Documentation10.1% (1,258)14.6% (1,825)9.6% (1,197)30.4% (3,791)35.3% (4,412)
Implementation3.9% (494)12.5% (1,566)9.0% (1,126)40.3% (5,056)34.3% (4,304)
Review10.0% (1,242)16.7% (2,083)8.8% (1,097)37.2% (4,630)27.3% (3,397)
Reference chart

Satisfaction with AI tools

When using AI tools or AI agents at work, which of the following have increased or decreased your usual satisfaction with completing the task? Select one answer for each.

All Respondents · n = 12,125

  • Greatly decreased
  • Somewhat decreased
  • No change
  • Somewhat improved
  • Greatly improved
  • Architecture exploration
  • Communication
  • Debugging/refactoring
  • Design
  • Documentation
  • Implementation
  • Review
View exact data
Percent and respondent count for All Respondents
ActivityGreatly decreasedSomewhat decreasedNo changeSomewhat improvedGreatly improved
Architecture exploration4.5% (530)4.8% (567)35.8% (4,207)31.9% (3,751)22.9% (2,686)
Communication10.0% (1,163)7.9% (927)49.5% (5,783)19.2% (2,239)13.5% (1,573)
Debugging/refactoring4.3% (511)6.2% (735)23.4% (2,781)35.7% (4,230)30.4% (3,604)
Design5.8% (682)7.3% (854)39.4% (4,636)28.7% (3,375)18.9% (2,221)
Documentation4.6% (545)5.2% (610)29.6% (3,489)27.9% (3,286)32.7% (3,858)
Implementation8.2% (976)11.7% (1,399)19.9% (2,377)33.6% (4,008)26.5% (3,161)
Review5.3% (623)6.6% (782)32.8% (3,856)32.3% (3,800)23.0% (2,707)
Reference chart

Did you switch AI tools in the past year?

In the past year, have you switched from one AI tool to another for work? If yes, what was the primary reason?

All Respondents · n = 13,144

  • No
    33.7%4,423
  • Yes — better output quality
    25.1%3,305
  • Yes — my organization required or recommended a different tool
    10.6%1,394
  • Yes — better integration with my workflow or development environment
    7.4%975
  • Yes — lower cost
    6.4%835
  • Yes — better support for agents, automation, or coding workflows
    3.7%487
  • Yes — another reason (write in)
    3.6%466
  • Yes — the previous tool became too expensive, limited, or unavailable
    3.3%433
  • Yes — better security, privacy, or compliance
    2.2%283
  • Yes — access to a preferred model or provider
    2.1%280
  • Yes — for a specific feature
    1.4%179
  • Not applicable — I do not use AI tools for work
    0.6%84
Reference chart

AI memory retention

When AI tools remember things about your work, who should control that memory?

All Respondents · n = 12,825

  • Just me
    53.6%6,877
  • My organization centrally
    14.7%1,891
  • I do not want AI to retain memory
    14.2%1,821
  • My team
    11.1%1,420
  • Not sure
    6.4%816
Reference chart

AI at work

Which of the following best describes how AI tools or agents are used at your organization today?

All Respondents · n = 13,167

  • AI use is mostly optional and left up to individuals
    29.7%3,915
  • My organization provides approved AI tools and guidance for how to use them
    24.0%3,154
  • Some teams actively encourage AI use, but practices vary across the organization
    15.9%2,097
  • My organization expects some employees to use AI tools for certain types of work
    11.5%1,507
  • My organization measures or evaluates AI use as part of employee or team performance
    10.2%1,344
  • My organization does not currently use AI tools or agents for work
    6.1%804
  • Not sure
    2.6%346
Reference chart

AI avoidance

Do you try to avoid using AI tools or agents at work or school? Select the option below that most accurately reflects your reasoning.

All Respondents · n = 13,857

  • I do not avoid using AI at work or school
    41.3%5,728
  • Concern over losing job skills or training an AI to replace you
    17.1%2,367
  • Moral or ethical concerns
    15.2%2,108
  • Issues with privacy or security
    13.0%1,802
  • Environmental concerns
    8.2%1,129
  • Other, please specify:
    2.7%369
  • Difficulty or lack of training
    2.6%354
Reference chart

Vetting new AI tools

Which of the following best describes how your organization currently approaches AI tools or models?

All Respondents · n = 12,637

  • We use a few approved tools or models
    32.8%4,138
  • We mostly use one approved tool or model
    18.6%2,345
  • Different teams choose their own tools or models
    17.3%2,186
  • We are trying multiple tools or models before deciding what to use longer-term
    12.5%1,576
  • We are not using AI tools or models in a meaningful way
    9.5%1,197
  • Not sure
    5.1%649
  • We are building internal tools or models
    4.3%546
Reference chart

Technology

3.1 Most popular technologies

Users using technology

In the last 12 months, has your role at work, school, or in personal projects intersected with any of the following types of settings involving technology? We are going to ask you about to ask you about tools you used within each setting next. Select all that apply.

Used less than other technologies surveyed this year, only 23% use technology related to security

All Respondents · n = 16,241

  • I used a programming, scripting, or markup language for any reason84.3%13,697
  • I used a development environment (IDE) to write code or interact with agent-assisted developer tools76.5%12,423
  • I used a database for storage, querying, deployment73.3%11,910
  • I used an LLM (large language model)71.8%11,655
  • I built, maintained or learned about web technologies62.7%10,176
  • I used a knowledge management platform (such as a wiki)62.0%10,074
  • I used a cloud platform or any tools related to cloud technologies59.5%9,659
  • I used a community platform or online public forum to search, comment, post to, or read from for technology-related content55.8%9,062
  • I used an enterprise search platform (a secure way to search for disparate internal content)24.6%3,987
  • I used security software for permissions and compliance operations23.1%3,755
  • None of these2.2%364
  • Other1.2%197

Subject areas listed here correspond to questions that follow

Reference chart

Programming, scripting, and markup languages

Which programming, scripting, and markup languages have you done extensive development work in over the past year, which do you want to work with over the next year, and which have you used AI tools or AI agents to work with? Select all that apply in each row.

Top used programming languages features JavaScript at the top of the list for 12th year in a row

All Respondents · n = 14,099

  • JavaScript62.0%8,742
  • SQL58.4%8,237
  • HTML/CSS58.2%8,212
  • Python58.0%8,184
  • Bash/Shell (all shells)51.3%7,231
  • TypeScript43.8%6,171
  • Java27.6%3,885
  • C#27.1%3,826
Show all 51 resultsShow fewer results
  • PowerShell24.6%3,472
  • C++24.3%3,425
  • C21.8%3,070
  • Rust18.0%2,542
  • PHP17.7%2,490
  • Go16.5%2,332
  • Kotlin10.8%1,524
  • Lua9.4%1,327
  • Assembly7.2%1,020
  • Swift5.9%831
  • Ruby5.8%823
  • Other5.2%733
  • Dart4.9%697
  • R4.5%639
  • Nix4.4%623
  • GDScript4.0%571
  • Visual Basic (.Net)4.0%557
  • VBA3.9%552
  • Groovy3.9%544
  • Perl3.8%538
  • MATLAB3.6%513
  • Zig2.8%398
  • MicroPython2.7%384
  • Lisp2.6%368
  • Elixir2.6%367
  • Objective-C2.5%357
  • Haskell2.3%324
  • Scala2.2%309
  • Delphi1.9%267
  • Fortran1.7%239
  • Erlang1.5%218
  • Julia1.5%211
  • Clojure1.3%187
  • Gleam1.2%176
  • Ada1.2%172
  • F#1.2%172
  • OCaml1.2%165
  • COBOL1.1%150
  • Prolog0.9%129
  • Solidity0.8%118
  • Mojo0.6%82
  • Odin0.2%34
  • Elm0.1%17
Reference chart

Programming, scripting, and markup languages

Which programming, scripting, and markup languages have you done extensive development work in over the past year, which do you want to work with over the next year, and which have you used AI tools or AI agents to work with? Select all that apply in each row.

All Respondents · n = 14,099

Technology use by survey year

Choose technologies

Select up to eight lines. All values are in the table below.

  • JavaScript
  • SQL
  • HTML/CSS
  • Python
  • Bash/Shell (all shells)
  • TypeScript
  • Java
  • C#
0%18%35%53%70%2017201820192020202120222023202420252026JavaScript · 2017 · 61.9%JavaScript · 2018 · 69.8%JavaScript · 2019 · 67.8%JavaScript · 2020 · 67.7%JavaScript · 2021 · 65.0%JavaScript · 2022 · 65.4%JavaScript · 2023 · 63.6%JavaScript · 2024 · 62.3%JavaScript · 2025 · 66.0%JavaScript · 2026 · 62.0%SQL · 2017 · 50.8%SQL · 2018 · 57.0%SQL · 2019 · 54.4%SQL · 2020 · 54.7%SQL · 2021 · 47.1%SQL · 2022 · 49.4%SQL · 2023 · 48.7%SQL · 2024 · 51.0%SQL · 2025 · 58.6%SQL · 2026 · 58.4%HTML/CSS · 2018 · 68.5%HTML/CSS · 2019 · 63.5%HTML/CSS · 2020 · 63.1%HTML/CSS · 2021 · 56.1%HTML/CSS · 2022 · 55.1%HTML/CSS · 2023 · 53.0%HTML/CSS · 2024 · 52.9%HTML/CSS · 2025 · 61.9%HTML/CSS · 2026 · 58.2%Python · 2017 · 31.7%Python · 2018 · 38.8%Python · 2019 · 41.7%Python · 2020 · 44.1%Python · 2021 · 48.2%Python · 2022 · 48.1%Python · 2023 · 49.3%Python · 2024 · 51.0%Python · 2025 · 57.9%Python · 2026 · 58.0%Bash/Shell (all shells) · 2018 · 39.8%Bash/Shell (all shells) · 2019 · 36.6%Bash/Shell (all shells) · 2020 · 33.1%Bash/Shell (all shells) · 2021 · 27.1%Bash/Shell (all shells) · 2022 · 29.1%Bash/Shell (all shells) · 2023 · 32.4%Bash/Shell (all shells) · 2024 · 33.9%Bash/Shell (all shells) · 2025 · 48.7%Bash/Shell (all shells) · 2026 · 51.3%TypeScript · 2017 · 9.4%TypeScript · 2018 · 17.4%TypeScript · 2019 · 21.2%TypeScript · 2020 · 25.4%TypeScript · 2021 · 30.2%TypeScript · 2022 · 34.8%TypeScript · 2023 · 38.9%TypeScript · 2024 · 38.5%TypeScript · 2025 · 43.6%TypeScript · 2026 · 43.8%Java · 2017 · 39.3%Java · 2018 · 45.3%Java · 2019 · 41.1%Java · 2020 · 40.2%Java · 2021 · 35.4%Java · 2022 · 33.3%Java · 2023 · 30.6%Java · 2024 · 30.3%Java · 2025 · 29.4%Java · 2026 · 27.6%C# · 2017 · 33.8%C# · 2018 · 34.4%C# · 2019 · 31.0%C# · 2020 · 31.4%C# · 2021 · 27.9%C# · 2022 · 28.0%C# · 2023 · 27.6%C# · 2024 · 27.1%C# · 2025 · 27.8%C# · 2026 · 27.1%
View exact data
Technology use by survey year, percent of respondents
Technology2017201820192020202120222023202420252026
JavaScript61.9%69.8%67.8%67.7%65.0%65.4%63.6%62.3%66.0%62.0%
SQL50.8%57.0%54.4%54.7%47.1%49.4%48.7%51.0%58.6%58.4%
Python31.7%38.8%41.7%44.1%48.2%48.1%49.3%51.0%57.9%58.0%
TypeScript9.4%17.4%21.2%25.4%30.2%34.8%38.9%38.5%43.6%43.8%
Java39.3%45.3%41.1%40.2%35.4%33.3%30.6%30.3%29.4%27.6%
C#33.8%34.4%31.0%31.4%27.9%28.0%27.6%27.1%27.8%27.1%
C++22.1%24.4%23.5%23.9%24.3%22.6%22.4%23.0%23.5%24.3%
C18.9%23.0%20.6%21.8%21.0%19.3%19.3%20.3%22.0%21.8%
PHP28.1%30.7%26.4%26.2%22.0%12.1%18.6%18.2%18.8%17.7%
Go4.2%7.1%8.2%8.8%9.6%11.2%13.2%13.5%16.4%16.5%
Lua2.8%————4.0%6.1%6.2%9.2%9.4%
Assembly4.9%7.4%6.7%6.2%5.6%5.5%5.4%5.4%7.1%7.2%
Swift6.4%8.1%6.6%5.9%5.1%4.9%4.7%4.7%5.4%5.9%
Ruby9.0%10.1%8.4%7.1%6.8%6.1%6.2%5.2%6.4%5.8%
R4.4%6.1%5.8%5.7%5.1%4.7%4.2%4.3%4.9%4.5%
Visual Basic (.Net)6.2%6.7%————4.1%4.2%4.4%4.0%
VBA4.3%4.9%5.5%6.1%4.7%4.5%3.6%3.7%4.2%3.9%
Groovy3.2%4.3%——3.0%3.3%3.4%3.3%4.8%3.9%
Perl4.3%4.2%—3.1%2.5%2.3%2.5%2.5%3.8%3.8%
MATLAB4.2%5.8%——4.7%4.1%3.8%4.0%3.9%3.6%
Objective-C6.4%7.0%4.8%4.1%2.8%2.4%2.3%2.1%—2.5%
Haskell1.8%——2.1%2.2%2.2%2.1%2.0%—2.3%
Scala3.5%4.4%3.8%3.6%2.6%2.6%2.8%2.6%2.7%2.2%
HTML/CSS—68.5%63.5%63.1%56.1%55.1%53.0%52.9%61.9%58.2%
Bash/Shell (all shells)—39.8%36.6%33.1%27.1%29.1%32.4%33.9%48.7%51.3%
Kotlin—4.5%6.4%7.8%7.0%9.2%9.1%9.4%10.8%10.8%
Rust——3.2%5.1%8.3%9.3%13.1%12.6%14.9%18.0%
Dart——1.9%4.0%6.0%6.5%6.0%6.0%5.9%4.9%
Elixir——1.4%—1.7%2.2%2.3%2.1%2.7%2.6%
Clojure——1.4%—1.9%1.5%1.3%1.2%—1.3%
Julia———0.9%1.3%1.5%1.2%1.1%—1.5%
PowerShell————10.8%12.1%13.6%13.8%23.2%24.6%
Lisp————1.3%1.3%1.5%1.5%2.4%2.6%
Delphi————2.1%3.3%3.2%1.8%2.5%1.9%
Erlang————0.8%0.9%1.0%0.9%1.5%1.5%
F#————1.0%1.0%1.0%0.9%1.3%1.2%
COBOL————0.5%0.7%0.7%0.7%1.0%1.1%
Fortran—————0.9%1.0%1.1%1.4%1.7%
OCaml—————0.6%0.7%0.8%1.2%1.2%
Solidity—————1.5%1.3%1.1%—0.8%
GDScript——————1.7%2.3%3.3%4.0%
Zig——————0.8%1.1%2.1%2.8%
Ada——————0.8%0.9%1.4%1.2%
Prolog——————0.9%0.8%1.1%0.9%
MicroPython———————1.6%2.3%2.7%
Gleam————————1.1%1.2%
Mojo————————0.5%0.6%
Reference chart

Code Testing tools

Which tools for code testing, reliability, and maintainability have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the tool and want to continue to do so, please check both boxes in that row.

ESLint is widely used in this survey due to its direct ties to JavaScript. Did you know that ESLint is an open source project with more than 1K contributors? See more at github.com/eslint/eslint

All Respondents · n = 10,604

  • ESLint37.1%3,935
  • Postman / Newman32.6%3,453
  • pytest28.0%2,972
  • Playwright24.9%2,642
  • JUnit21.4%2,274
  • Jest20.8%2,205
  • Vitest19.4%2,053
  • SonarQube / SonarCloud18.8%1,990
Show all 42 resultsShow fewer results
  • Sentry13.9%1,472
  • Selenium12.0%1,272
  • Datadog11.1%1,180
  • xUnit.net10.7%1,132
  • NUnit8.7%921
  • Cypress8.2%869
  • Testcontainers7.8%827
  • Other6.6%696
  • JaCoCo6.0%635
  • CodeRabbit5.7%605
  • BrowserStack5.4%576
  • Codecov5.2%553
  • Mocha5.1%545
  • CodeQL5.1%539
  • Snyk Code5.1%538
  • Cucumber4.8%508
  • k63.7%391
  • New Relic3.6%387
  • RSpec3.6%387
  • Semgrep2.4%252
  • TestNG2.3%240
  • Robot Framework2.1%226
  • Graphite2.1%223
  • Appium1.9%203
  • Canary1.5%162
  • Greptile1.2%130
  • DeepSource1.1%121
  • Qodo1.0%109
  • PHPUnit0.8%88
  • Sourcery0.8%88
  • Ellipsis0.8%82
  • Bruno0.6%60
  • GoogleTest0.5%56
  • MSTest0.4%47
Reference chart

Databases

Which database environments have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the database and want to continue to do so, please check both boxes in that row.

Ubiquitous Developer Adoption: Bryan Clark (Director of Product for LakeBase at Databricks) recalls observing this massive shift during his time at GitHub: "...inside GitHub, you could see every developer was switching to start using Postgres..." Listen to his interview on the Stack Overflow Podcast.

All Respondents · n = 12,061

  • PostgreSQL57.9%6,984
  • SQLite42.1%5,083
  • MySQL35.1%4,235
  • Redis28.3%3,414
  • Microsoft SQL Server27.1%3,270
  • MariaDB20.3%2,446
  • MongoDB19.0%2,297
  • Elasticsearch14.8%1,784
Show all 31 resultsShow fewer results
  • Oracle9.4%1,135
  • Dynamodb9.3%1,116
  • Supabase7.7%933
  • BigQuery6.7%809
  • DuckDB5.7%690
  • Valkey5.0%603
  • Cloud Firestore4.8%582
  • Other4.8%575
  • H24.7%564
  • Snowflake4.5%542
  • Cosmos DB4.3%513
  • Databricks SQL4.2%510
  • Firebase Realtime Database4.2%504
  • Microsoft Access4.0%483
  • InfluxDB3.6%432
  • Clickhouse3.6%429
  • Neo4J2.8%332
  • Cassandra2.5%300
  • IBM DB22.2%269
  • Amazon Redshift2.2%268
  • Pocketbase1.1%129
  • Cockroachdb1.0%117
  • Datomic0.5%64
Reference chart

Databases

Which database environments have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the database and want to continue to do so, please check both boxes in that row.

All Respondents · n = 12,061

Technology use by survey year

Choose technologies

Select up to eight lines. All values are in the table below.

  • PostgreSQL
  • SQLite
  • MySQL
  • Redis
  • Microsoft SQL Server
  • MariaDB
  • MongoDB
  • Elasticsearch
0%15%30%45%60%201820192020202120222023202420252026PostgreSQL · 2018 · 32.9%PostgreSQL · 2019 · 34.3%PostgreSQL · 2020 · 36.1%PostgreSQL · 2021 · 40.4%PostgreSQL · 2022 · 43.6%PostgreSQL · 2023 · 45.6%PostgreSQL · 2024 · 48.7%PostgreSQL · 2025 · 55.6%PostgreSQL · 2026 · 57.9%SQLite · 2018 · 19.7%SQLite · 2019 · 31.6%SQLite · 2020 · 31.2%SQLite · 2021 · 32.2%SQLite · 2022 · 32.0%SQLite · 2023 · 30.9%SQLite · 2024 · 33.1%SQLite · 2025 · 37.5%SQLite · 2026 · 42.1%MySQL · 2018 · 58.7%MySQL · 2019 · 54.0%MySQL · 2020 · 55.6%MySQL · 2021 · 50.2%MySQL · 2022 · 46.9%MySQL · 2023 · 41.1%MySQL · 2024 · 40.3%MySQL · 2025 · 40.5%MySQL · 2026 · 35.1%Redis · 2018 · 18.0%Redis · 2019 · 18.6%Redis · 2020 · 18.3%Redis · 2021 · 20.7%Redis · 2022 · 22.1%Redis · 2023 · 20.4%Redis · 2024 · 20.0%Redis · 2025 · 28.0%Redis · 2026 · 28.3%Microsoft SQL Server · 2018 · 41.2%Microsoft SQL Server · 2019 · 32.8%Microsoft SQL Server · 2020 · 33.0%Microsoft SQL Server · 2021 · 26.9%Microsoft SQL Server · 2022 · 26.9%Microsoft SQL Server · 2023 · 25.5%Microsoft SQL Server · 2024 · 25.3%Microsoft SQL Server · 2025 · 30.1%Microsoft SQL Server · 2026 · 27.1%MariaDB · 2018 · 13.4%MariaDB · 2019 · 16.5%MariaDB · 2020 · 16.8%MariaDB · 2021 · 17.2%MariaDB · 2022 · 17.9%MariaDB · 2023 · 17.6%MariaDB · 2024 · 17.2%MariaDB · 2025 · 22.5%MariaDB · 2026 · 20.3%MongoDB · 2018 · 25.9%MongoDB · 2019 · 25.5%MongoDB · 2020 · 26.4%MongoDB · 2021 · 27.7%MongoDB · 2022 · 28.3%MongoDB · 2023 · 25.5%MongoDB · 2024 · 24.8%MongoDB · 2025 · 24.0%MongoDB · 2026 · 19.0%Elasticsearch · 2018 · 14.1%Elasticsearch · 2019 · 14.3%Elasticsearch · 2020 · 13.8%Elasticsearch · 2021 · 13.3%Elasticsearch · 2022 · 12.2%Elasticsearch · 2023 · 13.4%Elasticsearch · 2024 · 12.5%Elasticsearch · 2025 · 16.7%Elasticsearch · 2026 · 14.8%
View exact data
Technology use by survey year, percent of respondents
Technology201820192020202120222023202420252026
PostgreSQL32.9%34.3%36.1%40.4%43.6%45.6%48.7%55.6%57.9%
SQLite19.7%31.6%31.2%32.2%32.0%30.9%33.1%37.5%42.1%
MySQL58.7%54.0%55.6%50.2%46.9%41.1%40.3%40.5%35.1%
Redis18.0%18.6%18.3%20.7%22.1%20.4%20.0%28.0%28.3%
Microsoft SQL Server41.2%32.8%33.0%26.9%26.9%25.5%25.3%30.1%27.1%
MariaDB13.4%16.5%16.8%17.2%17.9%17.6%17.2%22.5%20.3%
MongoDB25.9%25.5%26.4%27.7%28.3%25.5%24.8%24.0%19.0%
Elasticsearch14.1%14.3%13.8%13.3%12.2%13.4%12.5%16.7%14.8%
Oracle11.1%16.5%16.5%12.6%11.5%9.8%10.1%10.6%9.4%
BigQuery2.1%————4.5%4.8%6.5%6.7%
Cosmos DB7.9%————3.5%3.3%4.6%4.3%
Neo4J2.4%———2.1%1.9%1.9%2.7%2.8%
Cassandra3.7%3.5%3.3%2.7%2.7%2.5%2.4%2.9%2.5%
IBM DB22.5%—2.9%2.0%2.0%1.9%1.7%2.4%2.2%
Dynamodb—6.2%7.1%7.3%8.3%8.9%7.9%9.8%9.3%
Firebase Realtime Database—12.8%——8.7%6.4%5.8%5.0%4.2%
Cloud Firestore————7.5%6.4%5.4%5.7%4.8%
Supabase—————2.8%4.0%6.0%7.7%
DuckDB—————0.6%1.4%3.3%5.7%
H2—————3.7%3.8%5.0%4.7%
Snowflake—————2.4%2.6%4.1%4.5%
Microsoft Access—————4.3%4.2%4.8%4.0%
InfluxDB—————2.7%2.5%3.7%3.6%
Clickhouse—————1.1%1.7%2.4%3.6%
Cockroachdb—————1.0%1.0%1.0%1.0%
Datomic—————0.3%0.4%0.6%0.5%
Databricks SQL——————1.9%3.4%4.2%
Valkey———————2.4%5.0%
Amazon Redshift———————2.3%2.2%
Pocketbase———————1.0%1.1%
Reference chart

Cloud development

Which cloud platforms, containerization/orchestration tools, package managers, build tools, and infrastructure as code solutions have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the platform and want to continue to do so, please check both boxes in that row.

Cloudflare usage is up this year compared to last (25% vs. 20%) - is it due to the increase in bot activity which Cloudflare works to safeguard against?

All Respondents · n = 10,889

  • Docker / Docker Compose59.1%6,440
  • npm54.6%5,945
  • pip42.1%4,587
  • Amazon Web Services (AWS)41.8%4,549
  • Vite28.1%3,059
  • Homebrew27.3%2,978
  • Kubernetes26.0%2,836
  • Microsoft Azure25.1%2,737
Show all 57 resultsShow fewer results
  • Cloudflare24.6%2,680
  • Google Cloud24.5%2,663
  • uv23.3%2,534
  • Make22.8%2,479
  • pnpm19.0%2,065
  • NuGet18.6%2,026
  • Terraform18.3%1,998
  • Yarn16.7%1,823
  • APT16.6%1,813
  • Cargo16.3%1,770
  • Maven15.3%1,667
  • CMake14.9%1,626
  • Gradle14.8%1,611
  • Webpack14.7%1,603
  • Podman13.0%1,415
  • Helm12.4%1,352
  • Ansible12.0%1,302
  • Winget11.9%1,300
  • MSBuild11.8%1,286
  • Firebase11.1%1,204
  • Vercel10.7%1,166
  • Chocolatey9.7%1,053
  • DigitalOcean9.6%1,040
  • Composer9.2%1,002
  • Pacman9.1%995
  • Bun8.7%949
  • Hetzner Cloud7.8%849
  • Poetry7.4%801
  • Supabase6.9%756
  • Conda6.6%721
  • Nix5.7%622
  • Netlify5.6%615
  • Ninja5.3%573
  • RubyGems / Bundler4.6%503
  • Rancher4.4%480
  • Other4.2%459
  • Heroku3.6%397
  • OpenTofu3.6%388
  • OVHcloud3.3%361
  • Fly.io3.2%352
  • OpenShift3.2%345
  • Render3.0%330
  • Oracle Cloud Infrastructure (OCI)2.8%308
  • Bazel2.7%291
  • Railway2.6%282
  • Pulumi2.0%218
  • Vultr1.6%170
  • Scaleway1.4%151
  • IBM Cloud1.3%145
Reference chart

Web frameworks and technologies

Which web frameworks and web technologies have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the framework and want to continue to do so, please check both boxes in that row.

46% have used Node and 42% have used React, these are the top 2 web frameworks being used in this year's survey and last year's.

All Respondents · n = 10,281

  • Node.js46.4%4,769
  • React41.5%4,265
  • Next.js20.6%2,117
  • ASP.NET Core19.9%2,045
  • jQuery19.7%2,028
  • FastAPI18.5%1,903
  • Vue.js17.2%1,773
  • Angular16.0%1,642
Show all 41 resultsShow fewer results
  • Express15.6%1,604
  • ASP.NET13.9%1,428
  • WordPress13.3%1,365
  • Spring Boot12.8%1,314
  • Flask12.5%1,284
  • Django11.3%1,160
  • Svelte7.8%802
  • Laravel7.6%785
  • Blazor6.7%692
  • Astro6.2%634
  • AngularJS6.0%612
  • Other5.9%604
  • NestJS5.8%600
  • Ruby on Rails5.7%583
  • SvelteKit5.6%576
  • htmx5.4%555
  • Nuxt.js4.4%454
  • Deno4.0%416
  • Symfony4.0%413
  • Axum3.6%374
  • Fastify3.5%355
  • Hono2.8%287
  • Phoenix2.5%262
  • Drupal2.0%207
  • Preact1.9%193
  • Remix1.7%177
  • SolidJS1.5%155
  • Koa0.9%92
  • Quarkus0.3%26
  • Flutter0.2%24
  • Leptos0.2%22
  • Lit0.2%20
  • Lustre0.2%20
Reference chart

Web frameworks and technologies

Which web frameworks and web technologies have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the framework and want to continue to do so, please check both boxes in that row.

All Respondents · n = 10,281

Technology use by survey year

Choose technologies

Select up to eight lines. All values are in the table below.

  • Node.js
  • React
  • Next.js
  • ASP.NET Core
  • jQuery
  • FastAPI
  • Vue.js
  • Angular
0%15%30%45%60%2020202120222023202420252026Node.js · 2020 · 51.4%Node.js · 2021 · 33.9%Node.js · 2022 · 47.1%Node.js · 2023 · 42.7%Node.js · 2024 · 40.8%Node.js · 2025 · 48.7%Node.js · 2026 · 46.4%React · 2020 · 35.9%React · 2021 · 40.1%React · 2022 · 42.6%React · 2023 · 40.6%React · 2024 · 39.5%React · 2025 · 44.7%React · 2026 · 41.5%Next.js · 2022 · 13.5%Next.js · 2023 · 16.7%Next.js · 2024 · 17.9%Next.js · 2025 · 20.8%Next.js · 2026 · 20.6%ASP.NET Core · 2025 · 19.7%ASP.NET Core · 2026 · 19.9%jQuery · 2020 · 43.3%jQuery · 2021 · 34.4%jQuery · 2022 · 28.6%jQuery · 2023 · 22.0%jQuery · 2024 · 21.4%jQuery · 2025 · 23.4%jQuery · 2026 · 19.7%FastAPI · 2021 · 3.9%FastAPI · 2022 · 6.0%FastAPI · 2023 · 7.4%FastAPI · 2024 · 9.9%FastAPI · 2025 · 14.8%FastAPI · 2026 · 18.5%Vue.js · 2020 · 17.3%Vue.js · 2021 · 19.0%Vue.js · 2022 · 18.8%Vue.js · 2023 · 16.4%Vue.js · 2024 · 15.4%Vue.js · 2025 · 17.6%Vue.js · 2026 · 17.2%Angular · 2020 · 25.1%Angular · 2021 · 23.0%Angular · 2022 · 20.4%Angular · 2023 · 17.5%Angular · 2024 · 17.1%Angular · 2025 · 18.2%Angular · 2026 · 16.0%
View exact data
Technology use by survey year, percent of respondents
Technology2020202120222023202420252026
Node.js51.4%33.9%47.1%42.7%40.8%48.7%46.4%
React35.9%40.1%42.6%40.6%39.5%44.7%41.5%
jQuery43.3%34.4%28.6%22.0%21.4%23.4%19.7%
Vue.js17.3%19.0%18.8%16.4%15.4%17.6%17.2%
Angular25.1%23.0%20.4%17.5%17.1%18.2%16.0%
ASP.NET21.9%15.7%14.9%12.8%12.9%14.2%13.9%
WordPress14.2%——13.4%11.8%13.6%13.3%
Flask14.2%16.1%14.6%12.2%12.9%14.4%12.5%
Django14.2%15.0%14.7%11.5%12.0%12.6%11.3%
Laravel11.1%10.1%9.5%7.6%7.9%8.9%7.6%
Ruby on Rails7.0%7.0%5.8%5.5%4.7%5.9%5.7%
Symfony4.4%3.9%3.6%3.2%3.2%4.0%4.0%
Drupal3.2%2.4%2.2%1.9%1.9%2.2%2.0%
FastAPI—3.9%6.0%7.4%9.9%14.8%18.5%
Express—23.8%23.0%19.3%17.8%19.9%15.6%
Svelte—2.8%4.6%6.6%6.5%7.2%7.8%
AngularJS—11.5%9.0%7.2%6.8%7.3%6.0%
Next.js——13.5%16.7%17.9%20.8%20.6%
Blazor——4.5%4.9%4.9%7.0%6.7%
Nuxt.js——3.8%3.7%3.6%4.0%4.4%
Deno——1.7%2.4%1.9%4.1%4.0%
Fastify——1.9%2.1%2.2%2.9%3.5%
Phoenix——2.1%2.0%1.9%2.4%2.5%
Spring Boot———12.0%12.7%14.7%12.8%
NestJS———5.1%5.8%6.7%5.8%
Remix———1.3%1.6%—1.7%
Lit———0.7%——0.2%
Astro————3.0%4.5%6.2%
ASP.NET Core—————19.7%19.9%
Axum—————2.8%3.6%
Reference chart

LLMs

Which LLM model vendors for AI tools have you used for work in the past year? Which would you like to use next year? Select all that apply.

Most everyone uses Anthropic and OpenAI models, open-weight models from DeepSeek and Alibaba Qwen models are being used more, too.

All Respondents · n = 11,895

  • Anthropic74.6%8,871
  • OpenAI74.3%8,834
  • Google AI54.5%6,483
  • Microsoft26.4%3,142
  • DeepSeek18.9%2,253
  • Alibaba Qwen10.1%1,203
  • Mistral AI9.8%1,162
  • xAI7.6%906
Show all 25 resultsShow fewer results
  • Meta AI6.3%754
  • Moonshot AI4.0%479
  • Other3.6%427
  • Zhipu AI2.2%263
  • Snowflake2.0%234
  • Databricks (DBRX)1.6%194
  • Black Forest Labs1.0%120
  • Cohere0.9%112
  • Stability AI0.9%102
  • Voyage AI (MongoDB)0.8%92
  • Cascade0.7%82
  • 01.AI0.5%56
  • Luel0.3%41
  • Guide Labs0.3%39
  • Reka0.3%39
  • Minimax0.1%13
  • GLM0.0%4
Reference chart

Dev IDEs

Which development environments and code editors did you use regularly over the past year? Which do you want to use over the next year? Please check all that apply.

VS Code is still the most used IDE despite AI disruptors; usage has remained at or above 71% since 2021

All Respondents · n = 13,193

  • Visual Studio Code70.9%9,350
  • Notepad++27.8%3,673
  • Visual Studio27.6%3,643
  • Vim27.2%3,586
  • IntelliJ IDEA24.3%3,201
  • Cursor14.9%1,971
  • Neovim14.7%1,939
  • Android Studio14.2%1,880
Show all 36 resultsShow fewer results
  • PyCharm13.5%1,783
  • Nano12.1%1,600
  • Zed11.4%1,504
  • Jupyter Notebook/JupyterLab11.4%1,498
  • Xcode10.0%1,322
  • Sublime Text9.8%1,290
  • VSCodium7.7%1,017
  • Rider7.0%924
  • Eclipse6.9%910
  • WebStorm6.7%879
  • GitHub Codespaces6.6%876
  • PhpStorm5.6%743
  • Emacs5.3%697
  • CLion4.8%630
  • Other4.2%551
  • RustRover3.5%460
  • GoLand3.4%452
  • Windsurf3.2%423
  • Kiro2.8%365
  • Replit2.1%283
  • NetBeans2.1%280
  • Helix1.4%182
  • RubyMine1.4%180
  • Trae0.7%90
  • Antigravity0.4%58
  • R Studio0.4%52
  • Kate0.3%44
  • BBEdit0.2%29
Reference chart

3.2 Admired and Desired

Programming, scripting, and markup languages

Which programming, scripting, and markup languages have you done extensive development work in over the past year, which do you want to work with over the next year, and which have you used AI tools or AI agents to work with? Select all that apply in each row.

26% want to use Rust, 67% of Rust users want to continue using Rust

All Respondents · n = 14,099

  • Desired
  • Admired
  • Python
  • SQL
  • HTML/CSS
  • JavaScript
  • TypeScript
  • Rust
  • Bash/Shell (all shells)
  • Go
Show all 50 resultsShow fewer results
  • C#
  • C++
  • C
  • Java
  • Kotlin
  • PowerShell
  • Zig
  • PHP
  • Assembly
  • Lua
  • Swift
  • Elixir
  • Haskell
  • Nix
  • Ruby
  • Dart
  • GDScript
  • Lisp
  • R
  • Erlang
  • Gleam
  • OCaml
  • Julia
  • Clojure
  • Scala
  • MicroPython
  • F#
  • Perl
  • Ada
  • Visual Basic (.Net)
  • MATLAB
  • Fortran
  • Groovy
  • Mojo
  • Prolog
  • Objective-C
  • COBOL
  • Delphi
  • VBA
  • Solidity
  • Odin
  • Elm

Desired = "Want to use" / Admired = "Have used" & "Want to use"

Reference chart

Code Testing tools

Which tools for code testing, reliability, and maintainability have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the tool and want to continue to do so, please check both boxes in that row.

Bruno, PHPUnit, GoogleTest and MSTest may not be on the top of the desired list for most, but users of these testing tools rate them very highly on the admired scale

All Respondents · n = 10,604

  • Desired
  • Admired
  • ESLint
  • Playwright
  • pytest
  • Postman / Newman
  • Vitest
  • JUnit
  • SonarQube / SonarCloud
  • Sentry
Show all 41 resultsShow fewer results
  • Jest
  • Datadog
  • xUnit.net
  • Testcontainers
  • Selenium
  • CodeRabbit
  • NUnit
  • Cypress
  • BrowserStack
  • JaCoCo
  • Cucumber
  • CodeQL
  • Snyk Code
  • Codecov
  • k6
  • RSpec
  • Mocha
  • New Relic
  • Appium
  • Semgrep
  • Graphite
  • Robot Framework
  • Canary
  • TestNG
  • Greptile
  • DeepSource
  • Ellipsis
  • Sourcery
  • Qodo
  • PHPUnit
  • Bruno
  • GoogleTest
  • MSTest

Desired = "Want to use" / Admired = "Have used" & "Want to use"

Reference chart

Databases

Which database environments have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the database and want to continue to do so, please check both boxes in that row.

PostgreSQL is the most admired database technology this year, users love to use it

All Respondents · n = 12,061

  • Desired
  • Admired
  • PostgreSQL
  • SQLite
  • Redis
  • MySQL
  • Microsoft SQL Server
  • MongoDB
  • MariaDB
  • Elasticsearch
Show all 30 resultsShow fewer results
  • DuckDB
  • Supabase
  • Dynamodb
  • Valkey
  • BigQuery
  • Clickhouse
  • Oracle
  • Snowflake
  • Cassandra
  • Databricks SQL
  • Neo4J
  • Cosmos DB
  • Firebase Realtime Database
  • Cloud Firestore
  • H2
  • InfluxDB
  • Amazon Redshift
  • Cockroachdb
  • Microsoft Access
  • Pocketbase
  • IBM DB2
  • Datomic

Desired = "Want to use" / Admired = "Have used" & "Want to use"

Reference chart

Cloud development

Which cloud platforms, containerization/orchestration tools, package managers, build tools, and infrastructure as code solutions have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the platform and want to continue to do so, please check both boxes in that row.

All Respondents · n = 10,889

  • Desired
  • Admired
  • Docker / Docker Compose
  • Amazon Web Services (AWS)
  • npm
  • Kubernetes
  • Vite
  • uv
  • pip
  • Cloudflare
Show all 56 resultsShow fewer results
  • Homebrew
  • Google Cloud
  • Cargo
  • Microsoft Azure
  • Terraform
  • Make
  • pnpm
  • Podman
  • NuGet
  • APT
  • Ansible
  • Helm
  • Hetzner Cloud
  • CMake
  • Maven
  • Bun
  • Gradle
  • Pacman
  • Nix
  • Firebase
  • Winget
  • DigitalOcean
  • Vercel
  • MSBuild
  • Yarn
  • Supabase
  • Composer
  • OpenTofu
  • Chocolatey
  • Webpack
  • Ninja
  • Netlify
  • Conda
  • Fly.io
  • Rancher
  • Poetry
  • OpenShift
  • RubyGems / Bundler
  • Pulumi
  • OVHcloud
  • Railway
  • Oracle Cloud Infrastructure (OCI)
  • Render
  • Bazel
  • Heroku
  • Scaleway
  • Vultr
  • IBM Cloud

Desired = "Want to use" / Admired = "Have used" & "Want to use"

Reference chart

Web frameworks and technologies

Which web frameworks and web technologies have you done extensive development work in over the past year? Which do you want to work in over the next year? If you worked with the framework and want to continue to do so, please check both boxes in that row.

Svelte and SvelteKit are both highly admired, scoring 59% and 64% respectively

All Respondents · n = 10,281

  • Desired
  • Admired
  • React
  • Node.js
  • ASP.NET Core
  • FastAPI
  • Vue.js
  • Next.js
  • Angular
  • Svelte
Show all 40 resultsShow fewer results
  • Spring Boot
  • htmx
  • Django
  • Express
  • Flask
  • SvelteKit
  • jQuery
  • Astro
  • Blazor
  • ASP.NET
  • Laravel
  • Deno
  • Ruby on Rails
  • WordPress
  • NestJS
  • Axum
  • Nuxt.js
  • Phoenix
  • AngularJS
  • Symfony
  • Hono
  • Fastify
  • SolidJS
  • Preact
  • Remix
  • Drupal
  • Koa
  • Leptos
  • Lustre
  • Quarkus
  • Lit
  • Flutter

Desired = "Want to use" / Admired = "Have used" & "Want to use"

Reference chart

Dev IDEs

Which development environments and code editors did you use regularly over the past year? Which do you want to use over the next year? Please check all that apply.

IYKYK: Helix users want to keep using it as their IDE more than any other IDE this year

All Respondents · n = 13,193

  • Desired
  • Admired
  • Visual Studio Code
  • Vim
  • Notepad++
  • IntelliJ IDEA
  • Visual Studio
  • Neovim
  • Zed
  • PyCharm
Show all 35 resultsShow fewer results
  • Cursor
  • Android Studio
  • Jupyter Notebook/JupyterLab
  • Nano
  • VSCodium
  • Xcode
  • Sublime Text
  • Rider
  • WebStorm
  • Emacs
  • CLion
  • GitHub Codespaces
  • PhpStorm
  • RustRover
  • Eclipse
  • GoLand
  • Kiro
  • Windsurf
  • Helix
  • NetBeans
  • RubyMine
  • Replit
  • Trae
  • Antigravity
  • R Studio
  • Kate
  • BBEdit

Desired = "Want to use" / Admired = "Have used" & "Want to use"

Reference chart

LLMs

Which LLM model vendors for AI tools have you used for work in the past year? Which would you like to use next year? Select all that apply.

75% have used Anthropic LLMs in the last year, and 61% want to keep using them; 74% have used OpenAI LLMs, 42% want to keep using them.

All Respondents · n = 11,895

  • Desired
  • Admired
  • Anthropic
  • OpenAI
  • Google AI
  • DeepSeek
  • Microsoft
  • Mistral AI
  • Alibaba Qwen
  • Moonshot AI
Show all 24 resultsShow fewer results
  • xAI
  • Meta AI
  • Zhipu AI
  • Snowflake
  • Databricks (DBRX)
  • Black Forest Labs
  • Cohere
  • Voyage AI (MongoDB)
  • Stability AI
  • 01.AI
  • Cascade
  • Luel
  • Guide Labs
  • Reka
  • Minimax
  • GLM

Desired = "Want to use" / Admired = "Have used" & "Want to use"

Reference chart

3.3 Popular Knowledge Technologies

Knowledge Management Tools

Which knowledge management platforms have you used for work in the past year? Which would you like to use next year? Which have you used in your agentic workflows?

Confluence is the most used knowledge management tool this year

All Respondents · n = 9,142

  • Atlassian Confluence55.2%5,048
  • Google Workspace / Google Drive / Google Docs41.1%3,755
  • Microsoft SharePoint32.0%2,927
  • Notion18.6%1,700
  • Stack Internal / Stack Overflow for Teams7.0%644
  • Other6.6%602
  • ServiceNow Knowledge Management4.8%440
  • DokuWiki3.4%307
Show all 25 resultsShow fewer results
  • Obsidian2.8%252
  • XWiki1.9%174
  • Zendesk Guide1.6%143
  • BookStack1.4%131
  • Coda1.4%124
  • Outline1.1%101
  • Quip1.1%98
  • Document3600.9%79
  • Azure Devops0.8%74
  • Guru0.8%71
  • Gitlab0.7%67
  • Slab0.6%55
  • Mediawiki0.6%52
  • Nuclino0.5%44
  • Bloomfire0.4%34
  • KnowledgeOwl0.4%34
  • Helpjuice0.3%26
Reference chart

Enterprise Search

Which enterprise search platforms have you used for work in the past year? Which would you like to use next year? Which have you used in your agentic workflows?

Elastic is the most used enterprise search technology this year

All Respondents · n = 2,746

  • Elastic / Elasticsearch / Elastic Enterprise Search36.7%1,007
  • Microsoft Search22.3%613
  • OpenSearch17.4%478
  • Google Cloud Search / Vertex AI Search16.5%452
  • Azure AI Search / Azure Cognitive Search11.5%317
  • Apache Solr8.3%227
  • MongoDB Atlas Search7.9%218
  • Glean6.7%184
Show all 19 resultsShow fewer results
  • Algolia6.3%173
  • Other5.2%143
  • ServiceNow AI Search3.5%95
  • IBM Watson Discovery2.9%81
  • Amazon Kendra2.8%78
  • Lucidworks2.7%75
  • Salesforce Einstein Search2.7%75
  • Coveo1.6%43
  • Sinequa1.3%37
  • SearchStax1.3%35
  • Yext1.0%28
Reference chart

Security Operations Tools

Which security and permissions-based tools have you used for work in the past year? Which would you like to use, if you could choose, in the next year? Which have you used in your agentic workflows?

Auth0 and Okta are the most used Security Ops technologies this year

All Respondents · n = 4,052

  • Auth029.7%1,203
  • Okta28.7%1,162
  • Keycloak18.6%753
  • Sonar16.0%649
  • Datadog14.4%582
  • CrowdStrike12.4%502
  • Splunk10.1%409
  • Fortinet9.1%370
Show all 29 resultsShow fewer results
  • Other8.1%330
  • CyberArk6.4%260
  • SentinelOne5.6%228
  • Checkmarx5.0%203
  • BeyondTrust4.7%192
  • Wiz4.7%189
  • Ping Identity4.6%185
  • Mend.io3.2%129
  • Aqua Security3.0%123
  • Check Point3.0%122
  • OneLogin3.0%122
  • Teleport3.0%121
  • Veracode2.9%118
  • Tenable2.8%114
  • PortSwigger2.3%95
  • OpenText2.2%88
  • Delinea1.9%77
  • Orca Security1.7%68
  • Contrast Security1.0%39
  • Microsoft Entra0.8%32
  • authentik0.3%13
Reference chart

3.4 Build vs. AI

Did your team build tools?

Have you used or been involved with building in-house software in the past year? Are you planning to do so in the next year? Select all work functions from the list below that apply.

Only 19% of users building software are working to replace a vendor

All Respondents · n = 13,338

  • Small scripts/automations77.3%10,304
  • Team-level internal tools63.0%8,398
  • Customer-facing products54.4%7,254
  • Cross-team/platform tools46.0%6,130
  • Replacing paid vendors18.6%2,482
Reference chart

Did you build tools with AI assistance?

Which of the following, if any, has your team built with AI assistance in the past 12 months?

For those building software at work to replace a vendor, 54% say they used AI to help them

All Respondents · n = 10,338

  • Small scripts/automations74.2%7,640
  • Team-level internal tools66.5%5,588
  • Customer-facing products65.5%4,754
  • Cross-team/platform tools62.5%3,831
  • Replacing paid vendors54.4%1,349
Reference chart

Did AI cause business disruptions?

What kinds of AI mistakes have cost you or your team the most time in the past year? Select all that apply.

Security and privacy issues caused 24% of teams valuable time in the last year

All Respondents · n = 13,648

  • Plausible but wrong code69.3%9,452
  • Hard-to-review generated code54.2%7,403
  • Missing context49.3%6,724
  • Hallucinated package/config45.1%6,161
  • Incorrect or stale internal knowledge38.7%5,287
  • Outdated APIs33.9%4,621
  • Poor debugging path24.7%3,377
  • Security/privacy issue23.5%3,205
  • Licensing/source uncertainty11.4%1,560
  • Other8.6%1,172
Reference chart

Community

4.1 Finding and trusting answers

AI’s impact on Stack Overflow usage

How has AI changed your Stack Overflow usage? Select all that apply.

AI has fundamentally changed Stack Overflow's role as a resource for answers. 64% visit less for simple questions, 29% ask fewer questions, and 27% contribute less. Only 1.2% contribute more. However, 16% now visit more specifically for verification — meaning SO is shifting from "ask" to "verify." A small segment (6%) visit more because AI-generated answers lead them to SO for source attribution.

All Respondents · n = 16,985

  • Visit less for simple questions64.1%10,892
  • Ask fewer questions28.9%4,910
  • Contribute less26.5%4,496
  • No change21.8%3,704
  • Visit more for verification (for example, source attribution)15.9%2,707
  • Other6.4%1,084
  • Ask more advanced questions5.7%967
  • Visit SO more because AI-generated answers lead you there5.7%965
Show all 9 resultsShow fewer results
  • Contribute more1.2%203
Reference chart

Where technologists look for answers

Which actions do you turn to first when trying to find answers to a problem? Select your top 3 choices.

Online search (83%) and AI agents (70%) are the top two answer-seeking channels, both outpacing maintained repositories (44%) and trusted humans (29%). The implication: answer-seeking is now a self-serve, AI-augmented activity — human consultation is a third or fourth resort, not a first.

All Respondents · n = 16,664

  • Use online search tools to find general guidance and/or discourse82.7%13,774
  • Ask an AI agent69.9%11,643
  • Go directly to a maintained repository or more than one and search within its contents (this could be a repository like Wikipedia or IMDB or GitHub projects)43.7%7,288
  • Send a message to a human I trust directly28.8%4,803
  • Post a message in a group text or private discussion among humans I know14.8%2,462
  • Post in a public online forum8.0%1,327
  • Send a message to a human that I don't know directly1.7%276
Reference chart

What makes answers trustworthy

What makes you trust the information you find when you are searching for answers? Select your top 3 choices.

Trust is driven by presentation quality and social proof, not authority. Well-organized content (59%) and high engagement metrics — upvotes, likes, views (48%) — top the list, ahead of personal referral from a trusted human (47%) and citability/trackable origin (40%).

All Respondents · n = 15,355

  • The format and content of the information is well-organized and logical59.1%9,071
  • There are many upvotes/likes/shares/views associated with the content48.0%7,377
  • The information was referred to me by a human I trust directly46.6%7,159
  • The information has many citations, sources themselves don't matter as much as the ability to track down the origin39.8%6,111
  • I can see user activity associated with the human that posts the information I find25.3%3,890
  • I heard about the source of information before from another trusted source, such as a news site, a social feed, or a podcast interview24.3%3,730
  • The format and content of the information is creative and thought-provoking10.6%1,628
Reference chart

4.2 Communities and learning

Community participation

How do you participate in technical or professional communities? Select all that apply.

78% read discussions, but only 42% comment/react, 34% answer questions, and 31% ask questions. Open-source contribution sits at 28%, and event attendance at 21%. 14.5% don't participate at all. The community is a consumption economy: most people take far more than they give, and active contribution is a minority behavior.

All Respondents · n = 15,917

  • I read answers, posts, or discussions78.0%12,420
  • I comment or react41.8%6,660
  • I share resources with others35.2%5,597
  • I answer questions33.7%5,358
  • I ask questions31.2%4,972
  • I follow specific creators or experts29.7%4,734
  • I contribute to open-source projects28.1%4,468
  • I attend events, meetups, or webinars21.0%3,339
Show all 11 resultsShow fewer results
  • I participate in private groups or chats20.3%3,225
  • I do not currently participate in communities14.5%2,304
  • I moderate or help manage a community6.7%1,066
Reference chart

Barriers to community participation

What, if anything, prevents you from participating more in technical or professional communities? Select all that apply.

65% find answers without posting, and 45% say AI tools answer enough of their questions. Time is the third factor (36%). Beyond that, social friction is real: 21% fear judgment, 20% lack confidence asking publicly, and 12% feel communities aren't welcoming. Company-specific questions (23%) and privacy/legal restrictions (11%) also gate participation. Only 12% say nothing prevents them.

All Respondents · n = 15,602

  • I usually find answers without posting65.1%10,162
  • AI tools answer enough of my questions44.7%6,979
  • I do not have time35.8%5,588
  • My questions are too specific to my company or context22.6%3,532
  • I worry about being judged or criticized21.1%3,297
  • I do not feel confident asking questions publicly20.1%3,134
  • I prefer private conversations17.7%2,762
  • Nothing prevents me from participating more12.3%1,912
Show all 13 resultsShow fewer results
  • Communities do not feel welcoming to people like me12.1%1,893
  • Communities feel too advanced or specialized11.7%1,826
  • I have privacy, legal, or employer restrictions10.7%1,663
  • I do not know where to participate9.0%1,399
  • Not applicable2.7%425
Reference chart

Sources influencing skill development

Which sources most influence what technical, AI, or professional skills you decide to learn next? Select up to 3.

Skill direction comes from immediate social circles, not formal channels. Coworkers/peers (44%) and managers (33%) dominate. YouTube/video creators (26%) and online communities (23%) outpace documentation (18%) and job postings (20%).

All Respondents · n = 15,260

  • Coworkers, classmates, or peers43.9%6,695
  • Manager or workplace expectations32.7%4,986
  • YouTube or video creators25.8%3,931
  • Online communities or forums23.1%3,519
  • Job postings or labor market signals19.5%2,971
  • Documentation17.6%2,687
  • Newsletters or blogs13.5%2,062
  • AI tools or chatbots13.3%2,032
Show all 14 resultsShow fewer results
  • I am not currently deciding what to learn next11.0%1,673
  • Conferences or events9.4%1,434
  • Social media9.1%1,384
  • Courses or certifications7.1%1,077
  • Stack Overflow / Stack Exchange6.6%1,012
  • Podcasts5.5%843
Reference chart

How developers learned to code

How did you learn to code in the past year? Select all that apply.

Technical documentation (59%) is the #1 learning method, edging out AI codegen tools (53%) and general online resources (52%). Traditional methods — books (27%), colleague training (21%), school (14%) — are mid-tier. Bootcamps (6%) are the lowest formal option. The takeaway: developers are self-directed learners who combine docs, AI, and community resources rather than following a curriculum.

All Respondents · n = 14,530

  • Technical documentation (is generated for/by the tool or system)58.9%8,555
  • AI codegen tools or AI-enabled apps52.6%7,650
  • Other online resources (e.g. standard search, forum, online community)51.7%7,518
  • Videos (not associated with specific online course or certification)41.1%5,975
  • Stack Overflow or Stack Exchange33.8%4,909
  • Online courses or certification (includes all media types)29.3%4,260
  • Blogs or podcasts27.6%4,010
  • Books / physical media26.5%3,846
Show all 13 resultsShow fewer results
  • Colleague or on-the-job training21.4%3,106
  • School (i.e., university, college, etc)14.2%2,064
  • Games or coding challenges11.7%1,700
  • Coding bootcamp6.3%922
  • Other4.3%629
Reference chart

How developers learned AI programming

How did you learn to code for AI in the past year? Select all that apply.

AI codegen tools are the #1 method for learning AI itself (51%), followed by online resources (45%) and technical docs (43%). People are learning AI by doing AI.

All Respondents · n = 20,565

  • AI CodeGen tools or AI-enabled apps50.6%10,407
  • Other online resources (e.g. standard search, forum, online community)45.0%9,244
  • Technical documentation (is generated for/by the tool or system)42.5%8,733
  • Videos (not associated with specific online course or certification)37.8%7,781
  • Colleague or on-the-job training32.3%6,648
  • Blogs or podcasts29.5%6,069
  • Online Courses or Certification (includes all media types)17.1%3,516
  • Stack Overflow or Stack Exchange11.1%2,281
Show all 13 resultsShow fewer results
  • Books / Physical media7.7%1,583
  • Other6.3%1,290
  • School (i.e., University, College, etc)5.5%1,132
  • Games or coding challenges4.0%828
  • Coding Bootcamp3.8%777
Reference chart

4.3 Stack Overflow and community platforms

Stack Overflow account ownership

Do you have a Stack Overflow account?

77% of survey respondents have an SO account. This makes sense, we advertise this survey every year primarily on Stack Overflow.

All Respondents · n = 17,217

  • Yes77.4%13,320
  • No13.7%2,363
  • Not sure/can't remember8.9%1,534
Reference chart

Years using Stack Overflow

How long have you used Stack Overflow?

The user base is deeply tenured. 76% have used SO for 5+ years, with 34% at 10–15 years and 18% at 15+ years. Only 9% are new users (under 3 years), and 6% don't use it at all.

All Respondents · n = 17,142

  • Between 10 and 15 years33.5%5,739
  • Between 5 and 10 years25.0%4,291
  • More than 15 years, or since Stack Overflow started in 200817.5%3,000
  • Between 3 and 5 years9.1%1,561
  • I don't use Stack Overflow6.0%1,020
  • Between 1 and 3 years5.6%957
  • Less than one year3.4%574
Reference chart

Gender

Which of the following best represents your gender identity?

All Respondents · n = 21,149

  • Man86.0%18,191
  • Woman6.6%1,393
  • Prefer not to say4.3%899
  • Non-binary2.0%431
  • Prefer to self-describe:1.1%235
Reference chart

Learning new coding skills

Did you begin learning to code or learn a new coding skill or programming language in the last year?

It's nearly split down the middle — continuous learning is a norm for roughly half the developer population, while the other half is focused on applying existing skills.

All Respondents · n = 29,146

  • Yes52.0%15,169
  • No48.0%13,977
Reference chart

Learning AI skills

In the last year, did you spend time learning AI programming or AI-enabled tooling on your own or at work?

Self-directed learning dominates: 64% learn on their own time, with or without also learning at work.

All Respondents · n = 28,240

  • Yes, on my own23.4%6,607
  • Yes, at work15.4%4,360
  • Yes, both on my own and at work40.2%11,350
  • No21.0%5,923
Reference chart

Community platforms

Which community platforms have you used for technology-related content or purposes considerably or consistently in the past year? Which do you want to use over the next year? Please check all that apply.

GitHub (70%) and Stack Overflow (69%) are the top two platforms, followed by YouTube (58%) and Reddit (54%). The long tail includes Discord (33%), Medium (26%), Hacker News (24%), and X (16%). Emerging platforms like Bluesky (13%) are already ahead of Slack public channels (12%).

Worked with · n = 10,741

  • GitHub (public projects, not private repos)69.5%7,460
  • Stack Overflow68.6%7,372
  • YouTube58.4%6,272
  • Reddit53.8%5,784
  • Stack Exchange37.8%4,062
  • LinkedIn37.8%4,058
  • Discord33.4%3,591
  • Medium26.3%2,826
Show all 19 resultsShow fewer results
  • Hacker News24.0%2,576
  • X16.4%1,758
  • Bluesky12.6%1,356
  • Slack (public channels, not work)11.6%1,250
  • Dev.to10.3%1,103
  • Company-sponsored forums8.6%927
  • Substack8.0%858
  • Other, please specify6.7%719
  • Twitch5.3%569
  • Kaggle4.6%499
  • Hashnode1.2%131
Reference chart

Comments from Stack Overflow users about what changed about their habits on the site since AI took hold

In your own words, what changed about how you use Stack Overflow?

All Respondents · n = 5,350

I now refer to it for work/questions needing more nuance, or cases where I want to refer to prior art.

Developer, mobile, India, Stack Overflow user status is: Between 10 and 15 years

1 of 500
Reference chart

Knowledge

5.1 Context Challenges

Time spent finding information

In a typical week, how much time do you spend looking for information you need to do your work? Please provide a whole number representing hours per week, rounding up to the nearest whole number if needed.

75% spend 5 or more hours/week searching for information; 11% spend 20+ hours

All Respondents · n = 12,006

Respondents (%)0%5%10%15%20%16.7%6.2%9.8%16.1%510.5%17.6%106.9%157.1%2025304050+Hours per week

Scroll to see every hour value; hover a bar for its exact response count.

  • 1 hours: 2.6%, 316 respondents
  • 2 hours: 6.7%, 799 respondents
  • 3 hours: 6.2%, 747 respondents
  • 4 hours: 9.8%, 1,171 respondents
  • 5 hours: 16.1%, 1,935 respondents
  • 6 hours: 5.4%, 643 respondents
  • 7 hours: 1.8%, 216 respondents
  • 8 hours: 10.5%, 1,255 respondents
  • 9 hours: 0.4%, 47 respondents
  • 10 hours: 17.6%, 2,119 respondents
  • 11 hours: 0.0%, 5 respondents
  • 12 hours: 2.2%, 266 respondents
  • 13 hours: 0.1%, 10 respondents
  • 14 hours: 0.4%, 46 respondents
  • 15 hours: 6.9%, 832 respondents
  • 16 hours: 1.7%, 203 respondents
  • 17 hours: 0.0%, 6 respondents
  • 18 hours: 0.4%, 43 respondents
  • 19 hours: 0.0%, 1 respondents
  • 20 hours: 7.1%, 848 respondents
  • 21 hours: 0.1%, 10 respondents
  • 22 hours: 0.0%, 4 respondents
  • 23 hours: 0.0%, 5 respondents
  • 24 hours: 0.4%, 45 respondents
  • 25 hours: 1.0%, 119 respondents
  • 26 hours: 0.0%, 3 respondents
  • 27 hours: 0.0%, 1 respondents
  • 28 hours: 0.1%, 11 respondents
  • 29 hours: 0.0%, 1 respondents
  • 30 hours: 1.0%, 125 respondents
  • 32 hours: 0.1%, 11 respondents
  • 33 hours: 0.0%, 1 respondents
  • 35 hours: 0.1%, 16 respondents
  • 36 hours: 0.0%, 1 respondents
  • 38 hours: 0.1%, 7 respondents
  • 39 hours: 0.0%, 3 respondents
  • 40 hours: 0.4%, 46 respondents
  • 42 hours: 0.0%, 6 respondents
  • 43 hours: 0.0%, 1 respondents
  • 45 hours: 0.0%, 4 respondents
  • 48 hours: 0.0%, 6 respondents
  • 50+ hours: 0.6%, 72 respondents
Reference chart

Important project context

When working on a project or task, which types of context are most important for you to have? Select up to 5.

Project goals (85%) and code/technical docs (73%) are the most needed context types.

All Respondents · n = 12,678

  • Project goals or requirements85.2%10,806
  • Code, repositories, or technical documentation73.3%9,297
  • Prior decisions and rationale44.0%5,580
  • Related tickets, tasks, or issues36.9%4,679
  • Business goals or product strategy33.0%4,189
  • Pull requests, commit history, or release notes26.3%3,332
  • Customer or user feedback26.0%3,296
  • Team norms, preferences, or ways of working25.1%3,183
Show all 14 resultsShow fewer results
  • Compliance, security, or policy requirements19.7%2,498
  • Current status or ownership19.6%2,490
  • Past discussions in chat, email, or meetings15.8%2,000
  • Data, dashboards, or reports7.4%932
  • I do not usually need additional context2.1%266
  • Other0.6%79
Reference chart

Barriers to finding context

Which of the following make it difficult to get the context you need for work? Select all that apply.

Incomplete info (63%), knowledge in people's heads (61%), and outdated info (56%) are top barriers.

All Respondents · n = 12,374

  • Information is incomplete63.2%7,825
  • Important context lives in people’s heads60.9%7,541
  • Information is outdated56.5%6,989
  • Information is spread across too many tools53.4%6,603
  • I do not know where to look30.8%3,816
  • Important decisions are hard to find later26.1%3,226
  • Different teams have different versions of the truth25.4%3,144
  • Search does not return useful results23.3%2,881
Show all 16 resultsShow fewer results
  • AI tools do not have enough context20.0%2,471
  • I do not know which source is authoritative18.4%2,272
  • AI tools use the wrong context16.8%2,075
  • Access or permissions prevent me from seeing what I need15.8%1,961
  • I usually have the context I need9.6%1,193
  • I intentionally limit context because of privacy or security concerns6.1%753
  • Not applicable2.4%291
  • Other1.6%193
Reference chart

Managing work context

Which of the following do you do to manage or prepare context for your work? Select all that apply.

Context management is highly manual — keeping separate notes (54%) and starting fresh AI sessions (45%) dominate.

All Respondents · n = 12,039

  • Keep notes or prompts in a separate document53.9%6,495
  • Maintain personal documentation or notes53.8%6,482
  • Start new AI chats or sessions to reset context45.0%5,415
  • Maintain team documentation or wikis43.0%5,178
  • Ask teammates where to find information38.0%4,572
  • Manually select files or documents for AI tools34.7%4,174
  • Search across multiple tools33.2%3,993
  • Copy and paste information between tools32.7%3,939
Show all 12 resultsShow fewer results
  • Reuse saved prompts or templates15.7%1,892
  • Limit the context I share because of privacy/security concerns10.0%1,198
  • I do not do anything specific to manage context6.9%829
  • Other1.7%199
Reference chart

Discovering context too late

How often do you discover important work context only after you have already started or completed a task?

79% discover important context only after starting a task, highlighting systemic context gaps.

All Respondents · n = 12,687

  • Very often12.1%1,539
  • Often23.4%2,964
  • Sometimes43.1%5,465
  • Rarely16.8%2,136
  • Never1.0%130
  • Not sure2.0%253
  • Not applicable1.6%200
Reference chart

5.2 Knowledge sources and trust

Sources for work-related answers

When you need to answer a work-related question, which sources do you use most often? Select up to 5.

Coworkers (72%) and code repos (63%) are the go-to sources — internal beats external.

All Respondents · n = 14,462

  • Coworkers or teammates71.9%10,405
  • Code repositories or code comments62.7%9,065
  • Internal documentation (i.e. Design, product, or planning docs), wiki, or knowledge base60.1%8,687
  • Internal chat tools, such as Slack or Teams45.4%6,572
  • Official docs30.5%4,415
  • Product, vendor, or API documentation29.2%4,221
  • Stack Overflow24.0%3,472
  • Manager or leadership22.6%3,272
Show all 16 resultsShow fewer results
  • Project management or ticketing tools20.2%2,916
  • Past decisions or architecture records15.5%2,248
  • Meeting notes or transcripts10.6%1,536
  • Other7.0%1,008
  • Vendor forum2.8%411
  • Customer, support, or sales information2.5%362
  • Not sure1.4%196
  • None of these0.8%122
Reference chart

Importance of information accuracy

For each source, how important is it that information you use to operate is accurate?

Accuracy is the expectation for official docs, API documentation, and code repos; chat tools and Stack Overflow face lower accuracy expectations.

All Respondents · n = 14,049

Row values: Very important

  • Code repositories or code comments65.2%
  • Coworkers or teammates56.7%
  • Customer, support, or sales information50.6%
  • Internal chat tools, such as Slack or Teams35.0%
  • Internal documentation (i.e. Design, product, or planning docs), wiki, or knowledge base50.9%
  • Manager or leadership59.6%
  • Meeting notes or transcripts34.5%
  • None of these55.1%
Show all 15 resultsShow fewer results
  • Not sure43.0%
  • Official docs72.5%
  • Past decisions or architecture records38.4%
  • Product, vendor, or API documentation68.4%
  • Project management or ticketing tools41.2%
  • Stack Overflow30.0%
  • Vendor forum38.2%
Reference chart

Importance of current information

For each source, how important is it that information you use to operate is recent or clearly timestamped?

Recency/timestamps matter most for official docs and product/vendor docs; people tolerate stale human knowledge.

All Respondents · n = 13,678

Row values: Very important

  • Code repositories or code comments50.5%
  • Coworkers or teammates30.1%
  • Customer, support, or sales information49.0%
  • Internal chat tools, such as Slack or Teams35.8%
  • Internal documentation (i.e. Design, product, or planning docs), wiki, or knowledge base41.9%
  • Manager or leadership36.1%
  • Meeting notes or transcripts43.1%
  • None of these36.6%
Show all 15 resultsShow fewer results
  • Not sure27.8%
  • Official docs52.0%
  • Past decisions or architecture records35.4%
  • Product, vendor, or API documentation51.2%
  • Project management or ticketing tools44.5%
  • Stack Overflow33.7%
  • Vendor forum42.8%
Reference chart

Importance of source traceability

For each source you selected, how important is it that you can quickly check where the information came from before relying on it for work? By “check where the information came from,” we mean being able to quickly open or review the original source, context, citation, document, message, ticket, dashboard, or person behind the information.

Source traceability is most critical for code repos and customer support docs; coworker knowledge is trusted by reputation.

All Respondents · n = 13,214

Row values: Very important

  • Code repositories or code comments48.1%
  • Coworkers or teammates26.8%
  • Customer, support, or sales information44.2%
  • Internal chat tools, such as Slack or Teams31.5%
  • Internal documentation (i.e. Design, product, or planning docs), wiki, or knowledge base36.8%
  • Manager or leadership30.8%
  • Meeting notes or transcripts34.4%
  • None of these36.0%
Show all 15 resultsShow fewer results
  • Not sure25.0%
  • Official docs43.7%
  • Past decisions or architecture records36.5%
  • Product, vendor, or API documentation43.4%
  • Project management or ticketing tools38.5%
  • Stack Overflow32.7%
  • Vendor forum37.3%
Reference chart

Source attribution for AI answers

How important is source attribution when deciding whether to trust an AI-generated technical answer?

79% consider source attribution important or very important for trusting AI answers.

All Respondents · n = 13,945

  • Very important52.1%7,269
  • Important26.5%3,701
  • Somewhat important14.0%1,956
  • Not very important5.4%753
  • Least important1.9%266
Reference chart

Signals of trustworthy information

When deciding whether to rely on work-related information, which signals matter most to you? Select up to 5.

Recency (72.6%) and authority (60%) beat attribution and peer validation as trust signals.

All Respondents · n = 12,996

  • It is recent or clearly up to date72.6%9,435
  • It comes from an official or trusted source60.0%7,799
  • It is consistent with other trusted information45.9%5,967
  • It is clear, complete, and easy to understand42.6%5,540
  • I know who created or owns it38.1%4,946
  • It includes sources, citations, links, or supporting evidence37.6%4,887
  • It is marked final, approved, accepted, or official30.5%3,960
  • Other people on my team use or endorse it24.7%3,212
Show all 13 resultsShow fewer results
  • It was reviewed or validated by a subject matter expert24.3%3,161
  • It is not marked draft, outdated, deprecated, or superseded17.6%2,281
  • It respects access, privacy, or permission rules9.0%1,169
  • Not sure3.2%411
  • None of these0.9%115
Reference chart

5.3 AI adoption and cost

Tasks delegated to AI agents

For each job function, which tasks have you delegated to an AI tool or agent in the last 30 days?

Code tasks dominate AI use (73% write code, 62% debug); critical ops like deploying (10%) are rarely delegated.

All Respondents · n = 13,756

  • Writing, editing, or generating code72.9%10,027
  • Debugging or troubleshooting62.2%8,552
  • Writing or maintaining tests50.6%6,960
  • Writing or updating documentation48.3%6,642
  • Reviewing code, pull requests, or technical changes44.9%6,182
  • Making technical design or architecture decisions26.4%3,629
  • Investigating an incident, outage, or production issue25.9%3,560
  • Planning, scoping, or estimating work25.2%3,461
Show all 18 resultsShow fewer results
  • Answering questions from teammates, customers, users, or stakeholders21.4%2,941
  • Analyzing data, reports, or business metrics20.2%2,773
  • Onboarding to a new project, tool, system, or codebase19.7%2,706
  • Finding or fixing security, privacy, or compliance issues19.5%2,676
  • Changing production code, systems, or infrastructure18.9%2,601
  • Maintaining dependencies, packages, or development environments18.5%2,547
  • None of the above16.4%2,255
  • Monitoring systems, logs, metrics, or alerts13.6%1,864
  • Deploying or releasing software9.8%1,346
  • Coordinating work across teams or functions5.6%774
Reference chart

Control over information used by AI

When using AI tools or agents for work, how much control do you usually have over the work information they use?

64% actively curate what info AI uses; 17% have little or no visibility into AI inputs.

All Respondents · n = 13,243

  • I decide exactly what information to give the AI tool44.8%5,928
  • I choose from available sources, files, or systems19.3%2,549
  • I do not use AI tools or agents for work13.2%1,751
  • The AI tool chooses the information automatically10.4%1,374
  • I am not sure what information the AI tool uses6.3%830
  • Someone else manages what information the AI tool can access6.1%811
Reference chart

Factors in AI adoption

When you or your organization decides whether to adopt, continue using, or expand use of an AI tool or model for your work, which factors matter most? Select up to 3.

Output quality, security, and price drive adoption — organizations are pragmatic and risk-averse.

All Respondents · n = 11,462

  • The tool produces useful and accurate results30.7%3,520
  • The tool meets security, privacy requirements28.7%3,290
  • The price is acceptable for the value it provides27.0%3,090
  • The tool meets legal or compliance requirements22.3%2,557
  • Not sure22.0%2,522
  • Your company policy allows for usage16.1%1,844
  • Employees or teams are asking to use it12.7%1,451
  • There is evidence that the tool improves business outcomes12.6%1,446
Show all 18 resultsShow fewer results
  • The tool reduces time spent correcting, retrying, or reworking outputs12.2%1,401
  • The tool is worth paying more for because of better results9.9%1,137
  • Usage and spending are easy to monitor or limit9.5%1,089
  • The tool works well with how people already do their jobs9.2%1,055
  • The tool connects with systems people already use6.7%767
  • The organization can switch tools or models without major disruption5.8%665
  • Human oversight is manageable5.8%663
  • Other3.2%368
  • Review burden is low2.8%320
  • The tool has observability already integrated or built in1.9%213
Reference chart

Organizational AI usage practices

Which, if any, of the following has your organization done related to AI tool or model usage? Select all that apply.

38% of orgs set AI spending limits; 18% have no formal governance.

All Respondents · n = 11,468

  • Set limits on AI tool or model usage or spending38.1%4,367
  • Reviewed internal usage or spending data for AI tools or models28.3%3,240
  • Approved different AI tools or models for different types of work27.2%3,122
  • Set limits on who can use certain AI tools or models25.3%2,901
  • Increased spending on AI tools or models because results justified it20.6%2,366
  • Not sure17.8%2,043
  • None of the above17.7%2,030
  • Compared buying an AI tool with building something internally11.2%1,290
Show all 10 resultsShow fewer results
  • Stopped using or reduced spending on other software tools while increasing investment in AI tools or models8.7%992
  • Negotiated shorter commitments or the ability to switch AI tools or models over time3.2%364
Reference chart

AI cost and usage limits

Has cost or usage limits ever changed how you personally use AI tools at work?

33% self-monitor AI usage; cost pressure changes model selection and task selectivity.

All Respondents · n = 1,033

  • Yes, I personally monitor my own use32.6%337
  • Yes, I choose cheaper models when possible25.3%261
  • Yes, usage is metered to manage costs23.8%246
  • No, cost/limits have not affected my usage22.4%231
  • I have not encountered cost or usage limits20.6%213
  • Yes, I am more selective about tasks16.6%171
  • My company/team has increased budget for AI12.8%132
  • Not sure11.3%117
Show all 10 resultsShow fewer results
  • Yes, my company/team has switched models/vendors that are more affordable6.9%71
  • Yes, I choose faster models even when they cost more4.9%51
Reference chart

Cost-conscious prompting

Have you changed how you write prompts or structure requests to reduce cost or improve output quality?

75% changed prompts for quality, 38% for cost — quality optimization outpaces cost-cutting 2:1.

All Respondents · n = 560

  • Yes, to improve output quality75.4%422
  • Yes, to reduce cost38.4%215
  • No15.7%88
Reference chart

Data: Stack Exchange Inc. (2026) · ODbL 1.0 · Contents: DbCL 1.0