Jyotiraditya Gupta

ML Engineering Intern (Quant Research) at Connor, Clark & Lunn Investment Management

Conducted literature reviews and empirical validation to construct a multi-day forecasting signal using options chain for prediction of underlying securities, applying time-series modeling, Bayesian inference, and cross-sectional machine learning techniques. Engineered and backtested the predictive signals in R, using feature selection and multivariate lagged analysis to improve annualized risk-adjusted return by 0.5%.

ML Engineering Intern (Quant Trading) at Royal Bank of Canada, Capital Markets

Designed two alpha prediction signals using proprietary tick-level datasets, incorporating feature engineering, labeling strategies, and rigorous backtesting. Developed scalable data pipelines in KDB/Q, and Python for real-time market microstructure analysis, improving classification latency. Built execution signals leveraging order book imbalance, time-series signals, and statistical learning methods, increasing fill quality by 30% and Sharpe ratio by 87.5%. Reduced overnight risk from existing strategy by 6.7% and boosted strategy’s PnL by 22%.

AI Engineering Intern at Bell Canada

Built 15+ production-grade FastAPI services supporting downstream analytics and ML inference, processing 500K+ requests/day with 30% latency reduction.

AI Researcher (Algorithmic Fairness) at University of Toronto – Department of Computer Science

Working on clustering and summarization of large-scale opinion datasets with provable guarantees.

Machine Learning Research Assistant (LLM Compression) at University of Toronto – Department of Computer Science

Researched sparse and quantized compression techniques for DNNs and LLMs.