I study AI/ML from the ground up, starting from probability and working outward into information theory, statistical learning, and quantitative finance. My interest isn't in applying off-the-shelf models; it's in understanding the mathematical substrate that makes them work (or fail), and I work on the application of information-theoretic frameworks—such as entropy, mutual information, and coding theory—to financial markets and decision systems.
Currently moving through Feller's Probability as a rigorous foundation before Heard on the Street, and working through Polyanskiy & Wu's Information Theory in parallel, referring Stefan Moser's Lecture Notes.
I build and ship real tools,having currently made a production-grade NSE equity analytics desktop app.
| Book | Author | Why |
|---|---|---|
| An Introduction to Probability Theory and Its Applications | William Feller | Rigorous combinatorial & analytic probability — the foundation everything else rests on |
| Principles of Mathematical Analysis | Walter Rudin | The analysis and measure-theoretic bedrock beneath probability and finance |
| Elements of Information Theory | Cover & Thomas | Entropy, mutual information, and coding theory from the ground up |
| Information Theory | Yury Polyanskiy & Yihong Wu | Channel capacity, hypothesis testing, lossy compression — from first principles |
| Statistical Inference | Casella & Berger | Estimation and hypothesis testing done rigorously |
| Stochastic Calculus for Finance I & II | Steven Shreve | Martingales, Brownian motion, Black–Scholes via measure theory |
| Options, Futures, and Other Derivatives | John C. Hull | The practitioner's map of derivatives markets |
A production desktop app for end-of-day NSE equity analysis
Built with Polars for vectorised, out-of-core data processing. Features parallel downloads with retry/backoff, atomic writes, Camarilla pivot levels, multi-symbol comparison, and light/dark theming with full BaseWeb CSS coverage. Packaged as a Windows installer via PyInstaller + Inno Setup.
Python Polars Streamlit Plotly PyInstaller Inno Setup
A Python package for computing confidence intervals for binomial proportions
Implements a broad family of interval methods (Wald, Wilson, Clopper–Pearson, and more) with a clean, documented API. Published with hosted documentation on Read the Docs.
Python Statistics Confidence Intervals
A parser that extracts statistical moments and uncertainty from text
Turns unstructured/LLM output into structured statistical-moment representations for downstream analysis.
Python Statistical Moments Parsing
An open notebook tracking my progression through quantitative finance
Derivative pricing, stochastic calculus, and a structured roadmap from foundations to strategy.
Derivatives Stochastic Calculus Roadmap
Probability is the language of uncertainty. Learning to speak it fluently is the prerequisite for everything else.