Machine Learning Research Intern: Systematic Trading

We are looking for a Machine Learning Research Intern to work on predictive modeling problems in systematic trading.

The work sits at the intersection of machine learning, statistics, large-scale data analysis, and short-horizon market behavior. Projects are empirical and research-heavy: identifying useful structure in noisy data, building models that generalize, and understanding when a result is real versus an artifact of the research process.

Prior trading experience is useful, but not required. We care more about strong technical judgment, research ability, and the ability to build and test ideas rigorously.

What You’ll Work On

• Develop and evaluate machine learning models for short-horizon prediction.
• Work with large, high-frequency and intraday datasets.
• Explore new features, representations, and modeling approaches.
• Design experiments that are robust to leakage, overfitting, non-stationarity, and regime changes.
• Write clean, efficient, and reusable research code.
• Study model behavior across different instruments, time periods, and market conditions.
• Diagnose why models work, where they fail, and whether observed performance is likely to persist.
• Present results clearly and defend research conclusions with evidence.

What We’re Looking For

• Strong background in machine learning, statistics, applied mathematics, computer science, physics, engineering, or a related quantitative field.
• Meaningful research experience, whether through academic work, internships, publications, independent projects, or other technically rigorous work.
• Strong programming skills, especially in Python.
• Ability to work effectively with large and messy datasets.
• Good understanding of experimental design, validation, and model evaluation.
• Comfort working on open-ended problems where the correct formulation is not obvious.
• Ability to move between theory, implementation, and empirical analysis.

Useful Background

None of the following are strict requirements, but they are relevant to the work:

• Experience with time-series or sequential modeling.
• Exposure to systematic trading, market microstructure, or intraday market data.
• Familiarity with order flow, trades, quotes, or order book data.
• Experience with PyTorch, XGBoost, LightGBM, scikit-learn, Polars, NumPy, or similar tools.
• Experience with SQL, Linux, Git, distributed computing, or large-scale data systems.
• Familiarity with C++, Rust, Java, or another performance-oriented language.