Hetul Patel
ML & Quantitative Systems Engineer
I build systems at the intersection of ML infrastructure, quantitative research, and applied signal processing. My work is about hard tradeoffs: compression vs recall, overfitting vs generalization, model complexity vs interpretability. I measure everything, validate rigorously, and design for reproducibility.
Currently: AI agent governance (Aegisure) · Hyperspectral channel selection for remote sensing (AMT paper in prep) · Quantitative strategy research (GSIN)
Technical Focus
ML Infrastructure
Vector retrieval systems, model serving, distributed inference pipelines
Quantitative Research
Systematic strategy discovery, walk-forward validation, Monte Carlo simulation
Deep Learning & Fine-tuning
PyTorch, transformer fine-tuning, Hugging Face, LLM adaptation pipelines
Applied Signal Processing
Hyperspectral imaging, Gaussian spectral models, channel selection
AI Systems & Governance
Agent audit layers, policy enforcement, LLM tooling
Education
Master of Science in Computer Science
Auburn University at Montgomery
GPA: 4.0/4.0 · In Progress · Focus: Machine Learning, Algorithms, Systems, Quantitative Methods
Bachelor of Science in Computer Science
Auburn University at Montgomery
Skills
Programming
Python, C++, SQL
ML & Deep Learning
PyTorch, Scikit-learn, NumPy, Pandas, Hugging Face Transformers, LLM Fine-tuning, Vector Search, Retrieval Systems, Gaussian Process Models, Signal Processing
Systems & Backend
FastAPI, REST APIs, Distributed Systems, Redis, PostgreSQL, Docker, Supabase
Quantitative
Time-Series Modeling, Backtesting, Walk-Forward Analysis, Monte Carlo Simulation, Genetic Algorithms, Market Microstructure
Cloud & DevOps
AWS, Docker, CI/CD, Railway