Machine Learning Researcher

Who are we: Graviton is a privately funded quantitative trading firm striving for excellence in financial markets research. We trade across a multitude of asset classes and trading venues using a gamut of concepts and techniques ranging from time series analysis, filtering, classification, stochastic models, pattern recognition, to statistical inference analyzing terabytes of data to come up with ideas to identify pricing anomalies in financial markets.

As part of this team you will be tasked to apply machine learning and specifically deep learning techniques to trading problems while staying connected to broader research community. The researcher will put theory into practice and can immediately impact the global trading landscape with the expanding presence of Graviton in various markets.

Description

• Lead research in applying machine learning to a wide variety of datasets and trading problems
• Follow latest developments in academic research and incorporating research techniques from different fields of applications to our problems
• Improve tick-by-tick order book based time series feature sets using latest preprocessing techniques
• Work on current and develop new deep learning models to exploit large pool of in-house features and computing infrastructure
• Develop scalable pipeline for building predictive models across global markets
• Discover and implement new sources of predictive alpha, verify that they improve existing models, and integrate them into the firm's strategy development pipeline
• Partner with quant researchers and software developers in implementation of conducted research to production using Python / C++
• Advise infrastructure support team on latest developments on hardware and software to improve computing infrastructure for ML based research

Qualifications  

• Masters or PhD in Computer Science, Mathematics, Statistics, or a related field
• At least two years of demonstrated experience of ML/AI research in a professional setting or at a reputable academic institution
• Track record of academic publications preferred
• Experience with software engineering in Python / C++
• Experience with Tensorflow, Keras, PyTorch is highly desirable

Benefits:

This Machine Learning Researcher role develops and applies machine learning methods to quantitative trading problems. The work spans order-book time series, predictive models, and research pipelines, with research implemented in production alongside quantitative researchers and software developers.

Role at a glance

Skills mentioned

What you will likely work on

• Research machine learning approaches across datasets and trading problems.
• Evaluate developments in academic research and adapt useful methods to firm problems.
• Improve preprocessing and feature sets for tick-by-tick order-book time series.
• Develop deep learning models using in-house features and computing resources.
• Build scalable predictive-model pipelines for global markets.
• Identify potential predictive alpha, test its value, and integrate useful findings into strategy development.
• Work with quantitative researchers and software developers to bring research into production using Python or C++.
• Advise infrastructure support colleagues on hardware and software developments relevant to machine learning research.

Good fit signals

• You have at least two years of professional or academic ML/AI research experience.
• You hold a Master's or PhD in computer science, mathematics, statistics, or a related field.
• You have experience with software engineering in Python or C++.
• A track record of academic publications is preferred, not required.
• Experience with TensorFlow, Keras, or PyTorch is highly desirable.