AI/ML Engineer with 2 years of experience taking real AI products to production — agentic workflows, RAG pipelines, and MLOps built with LangGraph, LlamaIndex, and Google ADK. Here I showcase what I've shipped and share notes from building it.
Currently building
Early accessKnow what you eat. Reach your goals.
An AI calorie tracker: snap a photo of a meal and FoodVita estimates the food, portion, calories and macros — and shows how sure it is (every estimate carries its ±25% range), so nothing gets logged on a guess. Under the hood, a vision pipeline grounds macros against a 14.6k-food catalog; a local-first React Native app on an event-driven FastAPI backend, built solo.
Production AI systems — deployed, evaluated, and open source
Multi-domain RAG platform for grounded document Q&A
Upload documents across five domains — legal, financial, healthcare, research, technical — and ask questions answered strictly from retrieved chunks, with mandatory inline citations.
11-agent pipeline that automates the entire job application
Drop in a resume and a job posting; an orchestrated chain of agents parses both, runs gap analysis, and hands back ATS-optimized application materials — tailored, scored, and ready to send.
Thoughts on AI agents, machine learning, and production systems
Agentic engineering is the shift from AI that autocompletes lines to AI that plans, edits, tests, and ships across the whole SDLC. Here's what it actually is, the patterns that hold up in production, and how the engineer's job changes.
Frameworks, Tools, Platforms