Pushpjeet Cholkar

Cloud & AI/ML Consultant · Corporate Trainer · Indore, India

RAG & LLM Applications training (live)

Build a working retrieval-augmented application over real documents: chunking, embeddings, vector stores, prompt design, evaluation, and what breaks in production.

Live virtual · Intermediate · Needs Python basics · Fee on request

What you build

• End-to-end RAG over your own PDFs or document set
• Chunking strategies and why bad chunks produce confident wrong answers
• Embeddings and vector store choices without vendor theatre
• Prompt design that stays grounded in retrieved context
• Evaluation — measuring usefulness, not only answer similarity
• Production failure modes: cost, latency, permissions, prompt injection from retrieved text

Taught live by Pushpjeet Cholkar. Same engineering background as the AI/ML & RAG consulting work: seventeen years at the data layer for clients including Deutsche Bank, IBM, Wipro, Accenture, Capgemini, Cognizant and Oracle University.

Prerequisites

Python basics are required. You should be comfortable reading and writing simple scripts, installing packages and running code in a terminal or notebook. You do not need prior ML or LLM experience.

Corporate and open formats

Corporate cohorts can use your documents and cloud accounts. Open batches stay small enough that questions get answered in session — live, not recorded spam. Recordings are provided after each live class.

Related pages

• AI/ML & RAG consultant
• AWS Solutions Architect training
• AWS S3 — From Zero to Pro (batch starts 4 Oct 2026)
• pushpjeet.com — free AI fundamentals material and other live batches

Join or scope a cohort

Say whether you are an individual or booking for a team, and your Python comfort level. WhatsApp is fastest.