Nilesh Patil
AI systems & applied research. Head of AI at DreamStreet, building compliance-aware AI architecture for SEBI-regulated investor and trader workflows. Previously led applied AI research at Dream11 and ran a Columbia University research collaboration.
Particularly interested in AI harness design, developer productivity, and turning emerging model capabilities into reliable workflows and products.
More about me → · Publications → · Side projects → · CV → · Search →
Recent posts
-
Distributed K-Means Clustering in Python
May 20, 2020 · Implementing scalable k-means clustering using distributed computing frameworks -
Galactic Morphology using Deep-Learning
July 26, 2017 · Training a deep neural net to understand galactic structure -
Characterizing & analyzing networks : NYC taxi data
March 15, 2017 · Analyzing a real world graph : transportation network in NYC -
Working with numpy
March 4, 2017 · Vectors, matrices, and basic linear algebra with NumPy -
Human activity recognition
February 16, 2017 · Activity detection from sensor data -
Visualizing distributions
January 15, 2017 · Common visualization examples for distributions
All posts → · Posts by year → · Posts by tag →
Side projects
-
Data Science Docker Environment
July 1, 2018 · Opinionated Dockerfile + compose setup for a reproducible Python/R data-science stack: Jupyter, common ML libraries, ... -
Python vs Rust: k-means at scale
August 15, 2024 · Side-by-side performance comparison of k-means clustering implementations in Python (NumPy / scikit-learn) and Rust, ... -
AI/LLM Projects — In Progress
January 1, 2026 · Writeups for AI/LLM projects — agentic workflows, compliance-aware harness design, persona simulators, and evaluation... -
SimuCell3D: HPC tissue mechanics in C++
March 2, 2026 · Performance-oriented fork of an ETH-developed C++ framework for simulating 3D tissue mechanics at subcellular resolut...