About me.
About me.
I'm Sidharth Choudhary, a final-year dual-degree (MSc–MTech) student in Data and Computational Science at IIT Jodhpur, with a specialization in Mathematics. I build machine learning and AI systems that go the full distance — from the math on paper to code that runs in production. I'm most drawn to problems where the numbers carry real weight: forecasting market volatility, modeling risk, and lately, building agentic AI workflows that can actually reason through a task.
What I work on.
A few areas I keep coming back to:
Machine Learning & Data Science — end-to-end pipelines, from messy raw data to models that earn their place.
Quantitative Finance — volatility forecasting, risk modeling, and regime-aware strategies on real market data.
AI & Agentic Systems — exploring how LLMs and agent workflows can move beyond chat into systems that plan and act.
Mathematical Foundations — the part most people skip. I'd rather understand why a model works than just call .fit().
Selected Projects
Quant Volatility & Risk Engine
Volatility & risk system for Nifty 50 using GARCH-family models — regime detection, stress testing, and regime-aware allocation.
Real-Time Analytics & Insight Engine
Deployment-grade pipeline on UCI Online Retail II — cleaning, analysis, and an automated insight layer, served live via Streamlit + FastAPI.
SQL Analytics Dashboard
Industry-grade dashboard on a real relational dataset — advanced PostgreSQL: window functions, CTEs, and analyst-style query patterns.
Comparative Backprop for NMT
Optimizer comparison (SGD, Adam, RMSProp, Adagrad) on an English–Hindi Seq2Seq + LSTM + attention model. Best BLEU: 18.22 (RMSProp).