All programs

Generative AI & LLM Engineering

Build with the latest LLMs — RAG, fine-tuning, vector DBs, and AI agents.

Program fee
PKR 30,000
or up to 2 installments
Stack: LLM RAG LangChain OpenAI Vector DB Prompt Engineering Embeddings Agents Eval

Every evening at 9–11 PM (PKT), plus weekend deep-dives

No prerecorded lectures. Every session is live with a mentor — you ask questions, write code on the call, and get feedback in real time. Weekend sessions are extended deep-dives where you build the week's project end-to-end with the cohort.

Weeknight classes

Monday to Friday · 9–11 PM PKT. Concepts, walkthroughs, mentor Q&A. After your day job or classes.

Weekend deep-dives

Saturday & Sunday · extended hands-on sessions. Ship a working project every weekend with the cohort.

🎥 100% live · Zoom & Google Meet 📼 Recordings if you miss 💬 24h mentor reply on chat

Every product team needs someone who can ship AI features that actually work — not just chat demos. In 8 weeks you become the person on the team who knows how RAG, agents, fine-tuning, and evaluation actually fit together. Live sessions weeknights 9–11 PM PKT. Weekend builds are where you ship the week's AI feature end-to-end with the cohort — a chat-with-your-docs app, an agent that books meetings, a fine-tune you actually evaluate. We stay rigorous on cost, latency, and evaluation. The world has too many AI demos. We're producing engineers who ship.

Real projects · not toy exercises

⚖️

Model Evaluation Report

Honest side-by-side eval of 3 models on a real task.

🔎

Chat-with-your-docs App

A RAG pipeline with proper chunking, citations, and quality metrics.

🤖

Real Agent

A multi-step agent that books meetings end-to-end (no demo magic).

🛡

Eval Harness

A reusable eval suite to prove an LLM change is actually better.

🚀

Live AI Feature

A deployed, monitored AI feature with feedback collection.

If any of these sound like you, you're in the right place 👋

You're a software engineer wanting to be the AI specialist at your company.
You're a founder building an AI-first product and need real engineering depth.
You've hit the limits of "prompt engineering" and want to ship real systems.

What you'll learn

5 modules · 25 topics · hands-on the whole way.

01

LLM Foundations

  • How transformers work
  • Tokenization & embeddings
  • Comparing models
  • Cost & latency tradeoffs
  • OSS vs hosted models
02

Prompt Engineering

  • Prompt patterns
  • Few-shot examples
  • Output formatting
  • Evaluating prompts
  • Prompt versioning
03

Retrieval-Augmented Generation

  • Vector databases
  • Chunking strategies
  • Hybrid search
  • Citation handling
  • RAG eval
04

Agents & Tools

  • Tool use & function calling
  • Multi-step agents
  • Memory & context
  • Guardrails
  • Failure modes
05

Productionising AI

  • Streaming responses
  • Caching & cost control
  • Evaluation harnesses
  • Monitoring & feedback loops
  • PII & safety

Week by week, step by step

A clear path — not vibes. You'll know exactly what to ship at every checkpoint.

LLM fluency

Weeks 1-2
Deliverable: Side-by-side eval of 3 models

RAG pipeline

Weeks 3-4
Deliverable: Chat-with-your-docs app

Agents

Weeks 5-6
Deliverable: Agent that books meetings end-to-end

Ship to prod

Weeks 7-8
Deliverable: Public AI feature with monitoring

By the last week, you can…

🎯

Build production-grade RAG pipelines with proper chunking, retrieval, and citations.

🎯

Design agents that complete real multi-step tasks reliably.

🎯

Run rigorous LLM evaluations — no "looks good to me" benchmarking.

🎯

Control cost, latency, and quality in production AI systems.

Ready to start Generative AI & LLM Engineering?

Free forever. Mentor-led. Real projects. The kind of program you'd pay for — except you don't have to.

Other tracks

View all programs →