Software Engineering Intern
Built an internal work-order and preventive-maintenance platform for 70+ properties, 9,300+ rooms and 300,000+ sq ft using React/TypeScript, FastAPI, PostgreSQL, Docker and AWS.
MS Computer Science @ UT Arlington. I build RAG systems and AI agents, plus the evals and guardrails that keep them reliable when the data they read is hostile, on top of real-time ML infrastructure.
AI Engineering · RAG, Agents & ML Systems
I care most about the point where software meets reality: where a system either holds up under load, or under attack, or it does not. Lately that means AI systems: retrieval, tool-using agents, and the evals that prove whether they work.
The work I remember most is the work that broke first. Watching my own RAG agent email a secret to an attacker while showing the user a normal answer, then guarding it until 0 of 60 attacks got through. Profiling a model-induced p99 regression. Building an eval harness that caught my own agent reporting success too early.
The toolkit I bring into the match arena.
The 3D scene below borrows the striped oval outfield, red stands, floodlight towers and mountain wall of the Dharamshala ground.
Each red cricket ball represents a software project. Hover to identify it, then click to open the engineering story, tech stack, key result, and source repository.
Test-match timeline, session by session.
Built an internal work-order and preventive-maintenance platform for 70+ properties, 9,300+ rooms and 300,000+ sq ft using React/TypeScript, FastAPI, PostgreSQL, Docker and AWS.
Worked across PHP, modern web frameworks, SPA patterns, microservice architecture and Agile/Scrum delivery alongside Paycom engineers.
Built and deployed a full-stack application using Next.js, React, Firebase and Vercel, integrating Stripe, JWT auth, automated QA and regression testing.
Open to Summer 2027 AI/ML engineering internships, especially LLM applications, RAG and agent systems, evals and AI safety, or inference infrastructure.