Build Real AI Systems.
From Concepts to Production.
A 12-session, deeply hands-on course covering Generative AI, Large Language Models, Retrieval-Augmented Generation, and Autonomous Agentic Systems — taught by a software industry expert with over 25 years of experience.
What You'll Build & Learn
Every session produces a working artifact. You don't just study AI — you build it, test it, and understand why it works.
Understand LLMs & GenAI
Grasp how large language models think, hallucinate, tokenize, and respond — the mechanics behind ChatGPT, Claude, and Gemini.
Master Prompt Engineering
Move from ad-hoc prompts to systematic, production-ready prompt design using role priming, few-shot learning, and chain-of-thought techniques.
Build RAG Systems
Design and build Retrieval-Augmented Generation pipelines using embeddings, vector databases, and semantic search — the backbone of enterprise AI.
Create Agentic AI
Build autonomous AI agents that plan, reason, use external tools, and act with real agency — going far beyond simple chatbots.
Deploy to Production
Understand the infrastructure, scaling strategies, cost optimization, and monitoring required to run AI systems that serve real users.
AI Safety & Ethics
Apply bias mitigation, explainability, human-in-the-loop patterns, and responsible deployment practices from the ground up.
A Progressive, Project-Driven Journey
Every session builds on the last. You maintain a single running project — a hiring AI system — that evolves from a simple resume to a fully autonomous agent.
AI evolution, GenAI vs Agentic AI, LLM mechanics, tokens, temperature, context windows, and systematic prompt engineering.
JSON schema design, output parsing, pipeline construction, embeddings, vector databases, and building your first RAG system.
When and how to fine-tune models (LoRA, QLoRA), creating training datasets, rigorous evaluation, benchmarking, and A/B testing.
Scaling to production, cost optimization, AI safety, multi-agent architectures, tool use, memory systems, and the full capstone.
From Resume to Autonomous Hiring AI
Session 1 starts with creating a fictional student resume using ChatGPT. By Session 12, you'll have built a fully autonomous AI system that reviews, tailors, and finalizes resumes against job descriptions — completely on its own.
Lab Tests
Four primary lab tests assess your ability to build, not just describe. Each test is a real engineering challenge tied to the course's running project.
Generative AI & ChatGPT/Groq
Use Generative AI tools to produce structured, high-quality outputs. Explore prompt sensitivity and model behaviour across different platforms.
View Lab Details →AI Agents with n8n & Llama
Build your first autonomous AI agent using n8n workflow automation and Llama open-source models. Real automation, no hand-holding.
View Lab Details →End-to-End AI Application
Architect and build a complete end-to-end AI application integrating retrieval, generation, and structured outputs into a coherent system.
View Lab Details →Production AI System Design
Design a scalable, safe, production-ready AI system with a complete architecture review, deployment plan, and safety audit.
View Lab Details →AI Agents Library
Five working n8n agents built for this course — each a small, self-contained automation students import, configure, and run locally. Full source for every agent is on GitHub.
Good Morning Telegram Agent
The simplest possible n8n agent — 2 nodes, sends a daily good morning message to Telegram at 8 AM.
View Build Guide →Daily Weather Telegram Agent
Fetches live weather each morning and sends a formatted report to Telegram — no OAuth required.
View Build Guide →AI Joke Emailer
Click a button — Gemini AI writes a joke and it's emailed to your Gmail inbox seconds later.
View Build Guide →Birthday Reminder Telegram Bot
Reads a birthday list from Google Drive and only messages you on Telegram when there's a match.
View Build Guide →Stock News Morning Report Agent
Reads a stock watchlist from Google Drive, calls Gemini AI for news on each stock, and emails a full HTML report every weekday morning.
View Build Guide →With over 25 years in the software industry, Dinesh has built, shipped, and scaled technology products across global organisations. He brings deep engineering credibility to every concept — not as a researcher, but as someone who has navigated real production systems, led engineering teams, and coached executives. His teaching philosophy: you only truly understand AI when you've broken it and fixed it yourself.
AI for Teachers: An Open Textbook
Go deeper into the foundations, classroom applications, ethics, search, personalisation, and generative AI with this complete open textbook.
A linked reader for every chapter
Browse all 45 chapters and the additional essays from the AI4T project, then read the original book in a focused, in-page PDF reader.
Open Reference Reader →Ready to Build Real AI?
Start with Session 1 — no prerequisites beyond curiosity and a willingness to build.