Trinav Bhattacharyya

Software Engineer · Quantitative Developer · Distributed & Low-Latency Systems · AI/ML

Software Engineer with 3+ years of experience building and operating reliability-critical, distributed, and latency-sensitive systems at AWS Aurora. Promoted to SDE II for ownership of Aurora QA infrastructure and customer-impacting production reliability work. Experienced in C++, Java, Python, Linux, distributed databases, concurrency, and performance optimization.

Most recently, I worked in the Applied AI division at Sears, where I integrated Stripe payments and an MCP-compatible OpenAI chatbot into a React, Python, and MongoDB customer-support platform for AI-driven appliance diagnostics, repairs, and replacement orders.

I am currently pursuing an MS in Computer Science at Columbia University, specializing in Machine Learning and Artificial Intelligence, with an expected graduation in December 2026. My coursework includes Machine Learning, NLP, Advanced C++, Algorithms, Security, and Probability & Statistics.

My current research at Columbia's ARiSE Lab focuses on AI compliance, privacy, and Explainable AI. I am developing a constrained-decoding benchmark with 10,000+ scenarios to evaluate whether contemporary LLMs adhere to AI and privacy regulations. Previously, I worked on NSF-funded scientific machine learning research involving GPU-accelerated and differentiable simulations with Python and JAX.

Earlier in my research career, I developed open-source machine learning software and published 9 peer-reviewed papers with 400+ citations across IEEE, Springer, and Elsevier venues.

My interests include distributed and low-latency systems, quantitative software engineering, databases, concurrency, AI/ML systems, and security & privacy.