problyst.com/about

About the Instructor

Why this program exists, and who's behind it.

SN

Suman Nath

Staff AI/ML Engineer @ Intuit · RAG Pipelines · Agentic AI · Ex-Visa · Ex-TCS · Bengaluru

I've spent 14 years making complex systems reliable — the last 3 doing that for AI. At Intuit, I architect trustworthy agentic AI systems: rigorous evaluation frameworks, multi-agent architectures built on MCP (Model Context Protocol), and high-throughput knowledge pipelines using context-aware chunking and hybrid vector + graph retrieval (Milvus + Neo4j). Before that, at Visa, I built RAG pipelines for financial workflows with LangChain and an MCP-based triage agent connected to Splunk, ServiceNow, and Confluence in real time. And before that, 10+ years at TCS taught me what no architecture diagram can — leading 50+ engineers across India and the Netherlands, and staying calm when production goes down at 2am.

Here's the truth most courses won't tell you: building a RAG pipeline that demos well takes a weekend. Building one that handles thousands of concurrent users, respects data boundaries, invalidates stale knowledge, and explains its own failures — that takes a fundamentally different approach. I teach that difference. I love helping mid-to-senior engineers break into top-tier AI roles, and I write about applied AI, RAG systems, and the real cost of reliable intelligent systems at nathlabs.com.

I'm based in Bengaluru. If you're building production AI, evaluating RAG architectures, or deciding whether this bootcamp is right for you — connect with me on LinkedIn or book a call on Topmate. I'm always happy to talk.

principle/01

Live & Current

Recorded courses go stale in months. We teach live, with this month's tools — agents, evals, and the latest RAG tooling.

principle/02

Ship, Don't Watch

Every session ends with something working on your machine. You leave with a deployed app on your GitHub.

principle/03

Small Batches

Capped at 20 seats so every doubt gets solved live and every project gets real feedback.

Next batch — August 1, 2026

Want to learn together?

Join the next cohort or reach out with any question — replies within 24 hours.

See the Program