M.S. Applied Mathematics & Statistics at Columbia University. I build quantitative tools that people actually use: pricing and risk engines, data pipelines and LLM-powered analytics, shipped end to end from model to deployed app.
- 💼 Summer 2026: Summer Analyst, nonlinear rates desk at Crédit Agricole CIB
- 🔭 Currently building: open-source estimates of AI-lab revenue from public data
- 🧠 Interests: rates & volatility, market microstructure, applied ML/LLMs
- 📫 macisekander10@gmail.com
| Project | What it is | Stack |
|---|---|---|
| Open ARR Tracker · live | Open-source estimates of OpenAI and Anthropic annualized revenue from public disclosures and alternative data, with every input and assumption published | Python · GitHub Actions · Pages |
| Tranche Loss Simulator · live | Interactive capital-stack stress tester for GPU-backed structured credit | TypeScript · Next.js · Recharts |
| Fed Sentiment Analyzer | Scores FOMC minutes and Fed speeches as hawkish or dovish and tracks the Fed's stance over time | Python · Transformers · Docker |
| LLM Projects | Pretraining, fine-tuning, DPO, quantization, retrieval and multi-agent notebooks | PyTorch · Transformers |
| Machine Learning & Quant Finance | Regression, time series, neural nets, option pricing and CAPM performance attribution | scikit-learn · statsmodels · Keras |
- Languages: Python · TypeScript · SQL · R
- Quant: QuantLib · NumPy · pandas · statsmodels · arch · cvxpy
- ML / LLMs: PyTorch · Hugging Face Transformers · scikit-learn
- Apps & infra: Streamlit · Next.js · FastAPI · Docker · GitHub Actions · Vercel
- Data: Bloomberg (xbbg) · Databento · FRED · yfinance