Job Title

Singularity · Quantitative Trading · Full-time Paid Internship ABOUT THE ROLE Alpha begins with features. This role sits at the intersection of market microstructure and machine learning — your job is to translate raw market data into structured, signal-rich features that trading models can learn from. You will work across the firm's data stack, strategy research, and ML pipeline to build and validate the feature library that powers Singularity's quantitative edge in NIFTY 50 options and Indian derivatives. This is not a pure engineering role. You need to understand why a feature might carry information — and be rigorous about proving that it does. WHAT YOU WILL DO Design and construct features across three domains: technical indicators, market microstructure (order flow imbalance, bid-ask dynamics, volume profiles), and regime detection signals Build options-specific features — IV surface shape, skew, term structure, realized vs. implied vol spreads, open interest shifts Develop and maintain the firm's canonical feature library with versioning, documentation, and out-of-sample validation Run feature validation end-to-end: information coefficient (IC), decay analysis, correlation, and combinatorial purged cross-validation Collaborate with the research environment — implement and test strategies in notebooks, validate feature behaviour against backtest results Work with the AI research team to ensure features are correctly specified for model ingestion across the intelligence layer Identify and flag overfitting patterns, data leakage, and look-ahead bias throughout the feature pipeline REQUIRED Currently pursuing or recently completed a degree from A Tier 1 institute If above is not true, then a minimum of 1 year of relevant work experience Strong Python — pandas, NumPy, scikit-learn; comfortable with large time-series datasets Demonstrated understanding of financial ML — not just ML applied to finance, but the specific pitfalls: leakage, non-stationarity, low signal-to-noise Familiarity with Lopez de Prado's framework or an equivalent rigorous approach to financial feature construction Knowledge of derivatives market mechanics — how options price, vol surfaces, Greeks Statistical discipline: hypothesis testing, IC analysis, walk-forward validation GOOD TO HAVE Hands-on experience with NSE derivatives data Experience building factor models or signal research pipelines Familiarity with alternative data — FII/DII flows, options positioning data Research background — ability to read and implement academic papers Understanding of event-driven features: rollover dynamics, expiry effects