- Published on
- Ongoing
A running collection of the books, courses, hands-on practice, and tools I actually use for quantitative finance, deep learning, and working with large datasets. I keep adding to this as I find things worth sharing.
Mohit Appari
Welcome to my corner of the internet, where I dig deep into neural networks, the math and architecture behind them, quantitative methods, LLMs and AI agents, MCP servers, forecasting, and building production grade infrastructure.
I currently work as a Data Scientist at the Agency for Persons with Disabilities (State of Florida), with 3+ years of experience building machine learning, forecasting, and data-intensive systems across research and production, and I hold a Master's in Data Science from Florida State University.
My work spans training deep neural networks (CNNs, equivariant networks, transformers) and classical ML algorithms (regression, XGBoost, LightGBM) alongside time series and econometric modeling — ARIMA, GARCH, moving averages, and decomposition — to forecast spending, risk, and market behavior. Day to day that means analyzing millions of transactional records, building forecasts that land within 5–10% of actual spend, and engineering the ETL pipelines underneath them. I also build LLM-powered applications, RAG pipelines, and MCP servers, and I care a lot about turning research code into production systems that actually hold up.
This site is a collection of my thoughts, projects, resources, and everything I'm currently working on, the things I build and can't stop chasing.
Thank you for stopping by, feel free to reach out, happy reading!
Press ⌘ + K to start
Latest
- Published on
Notes on Andrej Karpathy's micrograd — a ~150-line autograd engine that builds backpropagation, gradient descent, and a full neural net (Value, Neuron, Layer, MLP) from first principles. Still the best free, intuitive explanation of how neural nets actually learn.- Published on
A deep dive into building a regime-aware trading system that uses Hidden Markov Models to classify market conditions and dynamically allocate across a strategy pool — from feature engineering to live paper trading.