$
Badri
Rajendran
Full-Stack Software Engineer · GenAI: Agents, RAG & Evals · 4+ yrs
Reliable AI is an engineering problem.
I'm a full-stack software engineer who treats it like one — multi-agent orchestration, RAG, and evals wired into end-to-end products people depend on, and owned from design through deployment and on-call.
@ Presenter Prep
with RAG
pricing (PolicyPal)
in production
01 readme
The engineer
behind the trace.
I'm Badri — a full-stack software engineer in the San Francisco Bay Area, with 4+ years building end-to-end web, mobile, and backend systems in Java, Python, and TypeScript. Recently I've been pulling generative AI out of the demo phase and into production: multi-agent orchestration, retrieval-augmented generation, LLM evals, and MCP servers, wired into customer-facing products and held to real reliability bars.
I care about the unglamorous parts — observability, cost tracking, test coverage, on-call — because that's the line between an AI feature that demos well and one people can lean on. I think in systems first: map the structure, stress-test it, then ship.
Today I'm a GenAI Engineer Intern at Presenter Prep in Mountain View, building LLM-as-Judge evals and a retrieval-grounded chatbot on a React/TypeScript and serverless Cloudflare stack. I hold an M.S. in Computer Science from Stevens Institute of Technology, and I'm open to Software Engineer, Full-Stack, and GenAI Engineer roles — teams building ambitious products where engineering rigor still matters.
// thinks in systems,
// ships in production.
Currently building PolicyPal — a cited-answer RAG assistant for US health coverage, now adding ACA plan comparison — and going deep on LLM evaluation & agent observability at scale.
What I'm good at
- Designing agentic systems that stay reliable, observable, and cost-aware in production
- Full product lifecycle ownership — API and schema design through CI/CD, deployment, and on-call
- Bridging classic backend engineering (Java/Spring, Python) with modern GenAI tooling
02 execution_trace
Where I've shipped.
Five engineering teams across two continents — each a node the trace runs through, most recent first.
03 artifacts
Things I've built.
From a flagship grounded-RAG assistant to civic-data ML — selected work, with source where it's public.
04 toolchain
What I work with.
A working toolkit, grouped by what it does — with the GenAI stack I reach for most up top.
05 training
Where I trained.
Stevens Institute of Technology
Master of Science in Computer Science
Anna University
Bachelor of Engineering in Computer Science
// signals & profiles
06 handshake
Let's build
something that holds up.
I'm open to Software Engineer, Full-Stack, and GenAI Engineer roles. If you're putting language models into a real product — or just want to talk agents, evals, and what makes them reliable — I'd love to hear from you.
badriathindran@gmail.com