Nilesh Patil
AI systems & applied research. I’m Head of AI at DreamStreet, building compliance-aware AI for SEBI-regulated (India’s securities regulator) investor and trader workflows. Before that I led applied AI research at Dream11 and ran a Columbia University research collaboration.
This site is where I write up the engineering behind that work - mostly field reports on HPC, applied ML, and the measurements behind the results.
About → · Publications → · Side projects → · CV → · Search →
Recent posts
- Adaptive OpenMP scheduling for tissue-mechanics simulations · 16 min read
- parallel Rust k-means - benchmarking a clustering rewrite against NumPy and scikit-learn · 19 min read
- Two-phase simplex in Python and Rust - a browser-trusted linear-programming solver · 18 min read
- Diffusion models - building an image generator from scratch, then the ladder to DALL·E 2 · 21 min read
- Grid pooling on a frozen backbone - Doppler angle estimation from B-mode ultrasound · 16 min read
- Hyperband on Dask - hyperparameter tuning for a variable-star classifier · 10 min read
All posts → · Posts by year → · Posts by tag →