AI Product Leader · Exited Founder · Building agentic systems with evals & safety at the core
-
I lead product for AI systems that have to work in the real world — agentic platforms, retrieval pipelines, eval frameworks, and the safety infrastructure that keeps them reliable and auditable at scale.
-
I'm an AI-native product leader who works hands-on with prompts, retrieval, and evals alongside my engineering partners — fast enough to prototype, deep enough to make sharp tradeoffs, careful about the line between product and engineering. I am passionate about building products that can work reliably in messy real-life situations and not just as impressive demos.
-
I have built this repositary to showcase some of the products and projects that I have or am currently working on. I have launched a couple of them comercially and use the rest to automate certain parts of my life. These are open source to help my fellow enthusiasts download, edit and use them to best fit their needs.
🔬 Levraging AI capabiliteis to bulid products that help increase productivity and automate parts of my life.
✍️ Writing about agentic systems, LLM evaluation, eval harnesses, and the practical realities of shipping AI, especially in regulated domains.
🎯 Exploring next opportunities in AI safety, evals, and agent quality
🤖 [Spellbound] — Launched commercially - LLM powered conversational product that gives medically credible (grounded in RAG) information and guidance to couples navigating fertility and pregnancy
🤖 [BlitzPrep] — Launched commercially - AI assistant to help find the right jobs, apply to them by tailoring your resume to fit the job being applied to and helping you prepare to maximize your chances of getting an offer (tailor it to fit your pace & your resume)
📊 [Brand OS] — Strengthen your brand on LinkedIn by creating and posting consistent content in your unique voice as well as improve it by tracking its performance
🛡️ [Trade X] — Deep equity research and automated trading agent
🧪 [RocketShop] — Create brand content for your D2C store
- Agentic systems — single-agent and multi-stage orchestration, planner/executor patterns, tool use
- LLM evaluation — LLM-as-judge, synthetic user cohorts, regression testing, online evals
- AI safety & reliability — guardrails, hallucination detection, confidence gating, human-in-the-loop escalation
- RAG & multimodal systems — retrieval grounding, OCR pipelines, structured extraction
- AI observability — telemetry, drift detection, execution tracing, rollback infrastructure
- Exited Founder & Head of Product at Spellbound (acquired Feb, 2026)
- Previously Senior AI PM at Magic Technologies and AI PM at KPMG India
- Carnegie Mellon University, MS in Software Management
- Based in San Francisco Bay Area