I am a software engineer with three years of experience building full stack applications, backend services, and ML systems. Completed my Master's in Computer Science at NYU.
I have worked at startups and unicorns deploying production systems using TypeScript, Node.js, Python, and C++. I enjoy building things that solve real problems and contributing to open source projects.
⚡ Fun fact — Music with code is love.
| Languages | Frontend | Backend and Data | ML and Cloud |
|---|---|---|---|
 |
|
Physics simulations using Vision Transformers and Joint Embedding models in PyTorch. Improved accuracy by 34% MSE over the ICLR baseline using distributed GPU training on NYU HPC. |
Entity, time, and hybrid retrieval over SEC filings using FAISS, BM25, and cross-encoder reranking paired with LLMs for financial question answering. |
|
Explored few-shot and in-context learning with LLaMA and Qwen models. Fine-tuned T5 for English-to-SQL conversion achieving an F1 score of 77%.
|
End-to-end ML pipeline for citrus leaf disease classification on GCP. Deployed inference on Kubernetes with model training, evaluation, and artifact management. |
|
MapReduce programs processing 10GB+ of NYC data using Hive and Apache Kafka for streaming analytics and insights.
|
Custom Java modules on the Floodlight controller for ARP handling, L3 routing, and round-robin load balancing across virtual IPs. |
|
Parallel mergesort and pi calculation benchmarks comparing thread scalability, overhead, and pinning behavior across C with OpenMP and Rust. |
A growing collection of data structures and algorithm solutions for interview prep and competitive programming. |
- Contributed to the Chromium project, integrating a next-generation display driver using C++ and Python
- Built VSCode tooling for the Linux organization for blockchain code analysis using TypeScript and Rust — 5000+ downloads