Building Solutions Leveraging Generative AI
Generative AI Products and AI Solutions Engineer
ideate, build, ship, learn, repeat
An AI agent that answers State Farm HR's workforce questions from structured HR data. Our LLM evaluation framework cut hallucinations and raised user satisfaction from 59% to 88%. UTD capstone.
Describe an event in one sentence; agents find venues, check logistics and draft the invites, then wait for your approval before anything is sent.
Checks a doctor's note against real Medicare coverage policies in under a minute: ready to submit, what's missing, or needs a clinician, with quotes from the note and policy. The AI never denies; a human confirms.
Ask a database questions out loud and get answers, no SQL needed.
Turning raw data into dashboards and decisions.
An end-to-end BI solution comparing EV charging demand with supply across Dallas–Fort Worth, with Tableau dashboards to guide where to invest.
A normalized relational database for player performance and scoring data, with SQL reporting views for analyzing trends and player stats.
A marketing campaign plan for Red Bull, from the product requirements to the analysis of how it should perform.
Products I've built with AI, live and ready to try.
See my career trajectory and get to know me in a fun, playable pixel-art game.
This portfolio website, with a RAG chat assistant that answers questions about my work from my resume. Using LLM function calling, Gemini decides on each question whether to answer or take an action, like opening my LinkedIn or GitHub.
Ask questions about my resume and get grounded answers, powered by retrieval-augmented generation.
Notes on building AI products: what worked, what didn't, and what I'd do differently.
I’ve spent the last few months doing two things at once: job hunting in the AI space, and building — trying to actually understand this technology instead of just reading about it.…
It started as a usual shopping activity. I was on Apple’s website picking out a new iPhone for my sister, and somewhere between comparing storage tiers and camera specs, I ended up…
I'm Hemanth, a determined, detail-oriented problem solver focused on Generative AI solutions engineering and agentic AI. I recently graduated from The University of Texas at Dallas with a Bachelor's degree in Computer Information Systems and Technology, and I enjoy turning real business problems into working AI products.
I build with LLMs, RAG, multi-agent workflows (LangGraph, MCP) and human-in-the-loop guardrails, and I care about the whole path: understanding how people actually work, prototyping fast, measuring quality with evals, and shipping something they use. That's what draws me to forward deployed engineering: working directly with teams to put AI where it improves decisions and efficiency.
I'm always up for a coffee and a conversation about the latest capabilities in technology and how to craft them into a solution for a real problem.
Open to roles and collaborations in generative AI. Email is the fastest way to reach me.