LLM-Ops & Generative AI Engineering
Become the engineer who ships production AI — RAG, agents and LLM-Ops that companies pay senior salaries for.
Our flagship technical diploma. Go from writing your first LLM call to deploying a monitored, production-grade AI system — the exact stack hiring teams are scrambling to staff in 2026.
- Level
- Intermediate → Job-ready · basic coding helpful
- Format
- Live online · evenings + weekend labs
- Duration
- 24 weeks · 180+ hours
- You leave with
- Prepares you for the Microsoft Azure AI Engineer Associate path
- Next intake
- Cohort 4 · 30 seats
Live online · evenings + weekend labs Outcomes, not just lecture notes.
Ship a production RAG + agent system end to end
Build an LLM-Ops pipeline with evaluation, monitoring and guardrails
Graduate with 4 deployable projects and a public GitHub portfolio
What you'll learn
Project-first and live. 11 modules built around something you ship and can show an employer.
Reserve your seat01 · Foundations of LLM Engineering
- Python for AI & the modern LLM API toolkit
- Tokens, embeddings & context windows
- Latency, cost and model-selection trade-offs
- Your first grounded LLM call in production shape
02 · Prompt Engineering as Code
- System & role design, few-shot and chain-of-thought
- Structured output (JSON / function schemas)
- Prompt versioning, regression tests & A/Bs
- Guarding against injection & jailbreaks
03 · Retrieval-Augmented Generation (RAG)
- Chunking strategies & embedding models
- Vector databases — pgvector, FAISS, Pinecone
- Hybrid search, re-ranking & citations
- Grounding answers and reducing hallucination
04 · Agentic Workflows & Tool-Use
- Function calling & tool orchestration
- Multi-step planning and state
- Graph-based agent flows (LangGraph-style)
- Human-in-the-loop & autonomy guardrails
05 · Fine-tuning & Adaptation
- When to fine-tune vs RAG vs prompt
- LoRA / PEFT on a budget
- Dataset curation & instruction tuning
- Evaluating a fine-tune honestly
06 · Evaluation, Testing & Safety
- Offline eval sets & LLM-as-judge
- Regression suites for prompts & chains
- Red-teaming and safety filters
- Measuring quality you can defend to a stakeholder
07 · LLM-Ops in Production
- Deployment patterns & serverless inference
- Tracing, monitoring & observability
- Caching, rate-limits and cost control
- Versioning, rollback and incident response
08 · Capstone — ship a real system
- Scope a production AI app end-to-end
- Build with eval + monitoring from day one
- Mentor architecture & code review
- Deploy to a public URL + GitHub you own
Build it. Ship it. Show it.
No slides-only theory — every week turns into something real you keep.
Every module ends in something you make — a working project, not a quiz.
You deploy and publish real work to a live link or repo you own.
You graduate with a portfolio you can put in front of an employer or client.
A practitioner, not a lecturer.
You learn the workflow your instructor actually runs in production — every session ends with something working, not a slide.
Rishi Aggarwal
Lead Faculty · LLM-Ops & GenAI Engineering
Former Staff Engineer at a fintech unicorn · shipped AI to 40M+ users
Rishi has spent a decade building ML systems that run in production — from real-time fraud models to RAG copilots used by millions. He teaches the exact loop his teams run: design, evaluate, deploy, monitor, repeat. No slides-only theory — every session ends with something running.
Lead faculty for this track
No vague promises — here's what's included.
Certificate of completion
Issued by LearnPact once your capstone ships
Azure AI Engineer pathway
Curriculum mapped to the Microsoft certification track
4 deployable projects
A public GitHub portfolio hiring teams take seriously
1:1 mentor reviews
Your code & architecture reviewed by faculty
Placement support
Interview prep, referrals and a hiring-partner loop
No-Cost EMI · scholarships
From ₹8,050/mo · need- & merit-based aid for those who qualify
Built to be applied, not just finished.
Where applied-AI talent gets hired
We're honest about who this is for.
We'd rather you join the track that actually fits than the loudest one. Here's the straight read.
- You can already write basic Python and want to build, not just understand, AI
- You're aiming to ship real AI features at work — RAG, agents, search, assistants
- You've tried a chatbot demo and now want to make one that holds up in production
- You're a developer or analyst ready to move into an AI engineering role
- You have 24 weeks to commit and want a portfolio of working, deployed projects
- If you've never written code, start with the ₹99 Sunday Series — free if money is tight
- If you want the AI fundamentals first, the 8-week Applied AI Foundations primer comes before this
- If you're still exploring whether tech is your path, try the free 2-minute path finder at /counsel
- If basic coding feels far off right now, Foundations will get you steady before this 24-week build
Not sure where you fit? Take the free 2-minute path finder →
Can't pay the full fee? You may not have to.
Need- & merit-based scholarships available for eligible applicants. No-Cost EMI from ₹8,050/mo keeps it manageable for everyone else. Not ready to commit? Try a ₹99 Sunday session first — free for anyone who needs it.
Ready to start LLM-Ops & Generative AI Engineering?
No fee to apply, and no obligation. Tell us about your background and goals — our admissions team replies within one working day with your seat, EMI options and scholarship eligibility.
- No fee to apply — and no obligation
- Fee Protection: 7-day full refund after your first paid session
- Scholarship assessment included
- Taught in English — explained in Bangla & Hindi when you need it
- A Sunday Series seat while you decide — free if you need it