I build scalable ML and backend systems—from distributed LLM training and RAG applications to production APIs serving high-volume workloads.
Based in Tempe, Arizona · Open to AI/ML infrastructure, ML systems, and backend engineering roles
- Improved distributed LLM training throughput by 30% on wafer-scale infrastructure at Cerebras Systems.
- Built backend services supporting 100K+ monthly merchant transactions with 99.9% uptime at Airpay.
- Optimized services and web experiences used by 500K+ active users at Jio Platforms.
| Project | What it demonstrates | Stack |
|---|---|---|
| EduSummarizer · Demo | Full-stack learning assistant that turns lectures and documents into grounded summaries | FastAPI, Next.js, RAG, FAISS, MongoDB, Cerebras |
| Notion Job Tracker | Chrome extension that extracts Greenhouse listings and saves them to a structured Notion workflow | JavaScript, Chrome APIs, Notion API |
| CampusQuest | Interactive 3D exploration experience for the ASU Polytechnic campus | Unity, C#, Mapbox |
- ML & AI: PyTorch, Hugging Face Transformers, LLM training and inference, RAG, model evaluation
- Backend & Data: Python, FastAPI, Node.js, Java, Spring Boot, PostgreSQL, MySQL, MongoDB, Redis
- Cloud & Infrastructure: AWS, Docker, Kubernetes, Terraform, distributed systems, CI/CD
- Observability: Datadog, Weights & Biases, performance profiling, reliability engineering
- Reliable and efficient LLM training and inference
- Evaluation and stress testing for agentic systems
- High-throughput distributed backend architecture