Context Engineering & RAG for PMs: A Live Build

Hosted by Rajesh Pentakota

Sun, Apr 19, 2026

5:00 PM UTC (1 hour)

Virtual (Zoom)

Free to join

Invite your network

Go deeper with a course

AI Coding for Product Managers: Build Real Products in 6 Weeks
Rajesh Pentakota
View syllabus

What you'll learn

The 4 context levers every AI-native PM has to own

Retrieval, chunking, re-ranking, and memory , what they are, when each one breaks, and the PM-level decisions

Build a working Agentic RAG pipeline

No boilerplate, no pre-built repos. Watch me go from 0 → 1 working demo over real Documents

Spot the silent failure modes that tank 80% of RAG Prototype

Hallucinated citations, retrieval drift, and stale memory, how each one looks in your product

A decision framework for when not to use RAG

Most PM teams use RAG as a hammer, let's double click and see when not to use it

Why this topic matters

PMs are sitting on top of a massive opportunity. As a PM turned solo‑founder, building CodePup AI from 0 to thousands of users has been the most fun and energizing chapter of my career, and this session is about sharing the tricks, tools, habits, and mindset that made that possible. This session will make you the person everyone bets on for the next big AI PM role!

You'll learn from

Rajesh Pentakota

Founder CodePup AI || ex-Zynga, Walmart, Flipkart

Rajesh has 15+ years of product leadership across Flipkart, Zynga, and Walmart (and more), where he led products at millions of DAU and millions of dollars in revenue. He’s now building in public as the solo founder of CodePup AI, an AI-powered eCommerce website builder.

He’s deeply hands-on with agentic AI and ships production systems across AI agents, RAG, reliable document parsing, long-context workflows, evaluation loops, and multi-agent orchestration, often building with Claude Code, Cursor, and Codex. linkedin

Previously at

Walmart
Zynga
Flipkart
Unbxd

Sign up to join this lesson

By continuing, you agree to Maven's Terms and Privacy Policy.