Live cohort 4 For engineers with 2+ years

From software engineer to AI engineer: Ship RAG, agents, MCP and voice AI in 12 weekends

Twelve weekends, live. You will not collect certificates. You will build six AI systems to a production bar, from a legal RAG bot to a voice agent that answers in under 800ms, and present your final project to the people who taught you.

Enroll now → See the 12 weekends

Cohort 4 begins Sunday, 25^th October. 22 live sessions, Saturdays and Sundays, 5:00 to 8:00 PM IST.

Inner circle price

US$660

Closes Tuesday, 20^th October. After that the price is US$840.

Includes the AI Engineering & Data Science Bootcamp 4.0 · Already own it? You pay only the difference · Enrollment closes Sunday, 1^st November.

12 weekends

Live, with the Diwali weekend off

25^th October to 16^th January

22 live sessions

Three hours each, Saturdays and Sundays

Recorded, with 1 year of access

6 projects

Graded builds, not tutorials

RAG, agents, MCP, voice, inference

1 showcase

Your final project, start to finish

Presented to mentors and peers

1.5M+

YouTube Subscribers

739K+

Learners

85K+

Paid learners

7000+

5 Star Reviews

The 12 weekends

Two versions of you exist twelve weekends from now.

Scroll. The line is what changes when you spend twelve weekends building AI systems instead of reading about them.

What you can shipWk 1

1 Week 1 · Kickoff

Your first LLM app.

The AI landscape, a fast Python refresh for engineers, APIs and decorators, and a call to a real model. By the end of day one you have something running.

Ends with: Assignment 0, an LLM app with a Streamlit UI

2 Weeks 2 to 3

RAG that holds up on contracts.

How LLMs and embeddings actually work, Qdrant, LangChain, Docling and chunking, then hybrid, SQL and graph RAG over legal and financial documents.

Ends with: Project 1, LegalBot with reranking and role-based access

3 Weeks 4 to 5

Agents that act on GitHub.

Tool calling and the ReAct loop, LangGraph workflows, memory, then subagents and handoffs, plus the eval harness that proves your system works.

Ends with: Project 2, a multi-agent DevOps system that triages CI/CD and reviews PRs

4 Weeks 6 to 7

Take it to production.

Multimodal AI, deployment on AWS AgentCore, MCP servers and clients, context engineering and agent harnesses with deepagents.

Ends with: Project 4, an IT helpdesk system: 2+ agents, an MCP tool, RAG and LangSmith tracing

5 Week 8

Know when to fine-tune.

Fine-tuning versus RAG, LoRA and QLoRA, synthetic datasets, small language models and Ollama, then your capstone is scoped and discussed.

Ends with: a fine-tuned small model running locally, and your capstone plan

6 Weeks 9 to 10

Build what a product team would ship.

AI product thinking, cost optimisation and OWASP Top 10 for AI, system design with stakeholders in mind, then a real-time voice stack.

Ends with: Project 5, a voice agent under 800ms or a multimodal agent

7 Weeks 11 to 12 · Final showcase

Serve it, route it, present it.

Inference engineering with vLLM, a router that sends each query to the right model, a live mock interview with Dhaval and Sid, and the final project showcase.

Ends with: Project 6, a cost and latency router, and your final showcase

What you leave with

Six projects that run, and one you present.

Each one comes with a spec a real team would hand you, and a bar it has to clear before it counts.

Project 1

LegalBot

Advanced RAG over contracts, filings and financial reports, with hybrid search, reranking and role-based access.

Project 2

DevOps agent

A multi-agent system for GitHub: one subagent triages CI/CD, another reviews PRs and acts on them.

Project 3

Eval harness

An offline eval set of 20+ examples, LLM-as-judge and at least one guardrail, added to your RAG or agent.

Project 4

IT helpdesk agent

A compound system: triage and resolver agents, an MCP tool to a ticketing system, RAG over runbooks, persistent memory.

Project 5

Voice or multimodal agent

A voice agent that answers in under 800ms end to end, or one that extracts data from three document types.

Project 6

Inference router

A small model served with vLLM, a router that sends hard queries to a larger model, and a cost report at 10K requests a day.

Final showcase

Your final project

Scoped with the instructors in week 8, built through the last phase, and presented to mentors and peers on 16th January.

The stack

The tools AI engineering teams actually use in 2026.

Python The language of every build

LangChain + LangGraph Chains, agents and workflows

Qdrant Vector search for RAG

Docling Document parsing and chunking

LangSmith Tracing and evals

MCP Servers and clients for tools

AWS AgentCore Deploying agents to the cloud

Ollama Running models locally

vLLM Serving models at scale

Unsloth Fine-tuning small models

LiveKit Real-time voice agents

JEV Building Decision Layer

Google Gemini Multimodal AI model

RAGAS RAG evaluation metrics

NeMo Guardrails LLM safety rails

Who is in the room

Taught by engineers who build these systems, not people who describe them.

Dhaval Patel

Founder, Codebasics · Ex-NVIDIA

Teaches the core build modules: LLM fundamentals, RAG, multi-agent systems, AI system design and the final project.

Hemanand Vadivel

Co-founder, Codebasics · Ex-Edgewell

Teaches the soft-skills track: AI product thinking, stakeholder management, LinkedIn strategy and personal branding.

Siddhant Pandey

AI Research Engineer, Codebasics

Leads the live weekend sessions and practice labs: hands-on coding, project walkthroughs, debugging and real-time problem solving.

Srikanth Dongala

AI Manager, AtliQ

Teaches practical AI engineering and system architecture through real-world projects and hands-on sessions.

12 weekends. 22 live sessions. 6 projects to a production bar.

Saturdays and Sundays, 5:00 to 8:00 PM IST. Open any week to see every session and what you build in it.

Kickoff & AI Fundamentals

Session 1 – Kickoff and Fundamentals

AI landscape · Python fundamentals · Simple functions · File manipulation · Classes and objects · APIs and decorators · Calling the Groq API

LLMs, Embeddings & RAG Foundations

Session 2 – LLMs, Embeddings & Transformer Architecture

How does an LLM work? · Transformer Architecture · Embeddings & Semantic Similarity · Key Parameters in LLMs

Session 3 – Vector DBs & RAG

Introduction to VectorDB · Qdrant Hands-on · RAG in Pure Python · Qdrant Advanced Operations

LangChain & Advanced RAG

Session 4 – LangChain, Docling & Chunking Strategies

LangChain Intro · RAG in LangChain · Docling Document Parsing · Hierarchical Chunking

Session 5 – Advanced RAG Techniques & Hands-On

Vector & Vectorless RAG Architectures · Hybrid RAG · SQL RAG · Graph RAG

Agentic AI and LangGraph

Session 6 – Agentic AI Foundations

What are AI Agents? · Tool Calling & ReAct Loop · Building your first Agent with LangChain · Routing with Semantic Router & JEV · Memory in Agents

Session 7 – Agent Orchestration & LangGraph

Introduction to LangGraph · Workflows · Conditional branches and loops · A ReAct loop in LangGraph · Hands-on

Multi-Agent Systems & Evals

Session 8 – Multi Agent Systems

Multi-agent systems · Subagents and handoffs · Multi-agent architectures · Subagent design for CI/CD triage and PR review

Session 9 – Evals, Guardrails & Observability

Online versus offline evals · Evaluating RAG pipelines · Evaluating agents · Guardrails · How it all comes together

Multimodal AI & AWS Deployment

Session 10 – Multimodal AI

Architecture of a multimodal AI model · Extract data from PDFs/images · A multimodal RAG System

Session 11 – Deployment on AWS

Introduction to AWS's Agent Stack · Deploying LangChain on AgentCore · AgentCore Services

MCP, Context & Harness Engineering

Session 12 – MCP

What is Model Context Protocol · MCP Architecture · Building MCP Servers · Building MCP Clients

Session 13 – Context Engineering & Harness Engineering

Why context matters in complex AI systems · Context engineering fundamentals · Context propagation in agents · Harness engineering · Harness patterns for tool use, memory and recovery · Hands-on with deepagents

Fine-Tuning & Capstone Discussion

Session 14 – Fine Tuning & Local AI

Fine-tune vs RAG: when to opt for what? · LoRA & QLoRA training · Synthetic dataset pipelines · What are Small Language Models? · Ollama

Session 15 – Capstone Project Discussion & Assignment Showcase

Assignment showcase and capstone scoping

AI Product Thinking & Security

Session 16 – AI Product Thinking and Reinforcement Learning

Turning AI capabilities into valuable products · Thinking like an AI product builder · The intuition behind reinforcement learning · Reward design

Session 17 – Advanced AI Engineering & Security

Cost optimization at scale · Semantic caching · Smart model routing · OWASP Top 10 for AI · Defending your system from adversarial attacks

System Thinking & Voice AI

Session 18 – Stakeholder Management and System Design for AI

Managing stakeholder expectations · How an AI system is different · AI system design patterns · Production caveats

Session 19 – Voice AI

The real-time voice stack · Low-latency voice systems · Interruptions and silence handling · Multi-turn context recovery

Inference Engineering & Personal Branding

Session 20 – Inference Engineering

Runtime optimisations: batching, caching, quantization, speculation · Infrastructure: routing, load balancing, autoscaling · Serving an LLM endpoint with vLLM

Session 21 – Personal Branding & Mock Interview

LinkedIn & Twitter positioning · GitHub portfolio storytelling · Live Mock Interview with Dhaval Patel & Siddhant Pandey

Final Project Showcase

Session 22 – Final Project Showcase

Present your final project to mentors and peers

What learners say

Verified reviews from the engineers who went before you.

Every review below is published by a real Codebasics learner on our platform.

AI Engineering cohorts 878 engineers

Across cohorts 1 to 3, from engineers with two years of experience to principal engineers and architects.

Dinakara Shetty

Cohort 1

★★★★★

As a software engineer with over two decades of experience, stepping into a new domain can sometimes be challenging, but my experience with the Codebasics AI bootcamp has been phenomenal. The instructors broke down complex concepts into explanations that were incredibly easy to understand. What really set this course apart was the focus on hands-on experience. Writing the capstone project completely from scratch solidified my learning in a way theory alone never could. I also really appreciated the soft skill sessions, which provided an unexpected but highly valuable takeaway.

AI Engineering

Sunny Dubey

Cohort 2

★★★★★

It covers some fairly complex AI engineering concepts, but they are explained in a simplified and practical way. More importantly, there is a strong focus on actually implementing what we learn, rather than just understanding the theory. We had people with 2+ years of experience, 5+ years of experience, professionals with 15+ years including solution architects, people already working with AI, and some who were completely new to AI. That mix made the discussions much more interesting.

AI Engineering

Asish Kumar Mishra

Cohort 2

★★★★★

The course is thoughtfully structured and delivers real value. If you follow it with the right intent and rigour, the transition to AI Engineering genuinely becomes easier. What I appreciate most: there's no spoon-feeding. They teach the fundamentals and give you the right pointers, so you learn to figure things out yourself, exactly the skill an AI engineer needs.

AI Engineering

Monish Mamilla

Cohort 1

★★★★★

As a software engineer, I especially liked how the program was organized step by step, starting from LLM fundamentals and embeddings, then moving into RAG, agents, guardrails, evaluations, LangGraph, observability, and deployment. What makes this bootcamp different is the strong focus on real-world AI engineering instead of only theory.

AI Engineering

Manal Kariapper

Cohort 1

★★★★★

The course content was very well structured, and the instructors were highly effective at explaining the core concepts while also sharing valuable nuggets of wisdom. The program kept me accountable through relevant project assignments, which helped build a real sense of accomplishment.

AI Engineering

On video Watch

Hear it from them directly. Each one opens their full video review.

AI Engineering Cohort 4 Opens 16^th Jan

Reviews open after the final showcase on Saturday, 16^th January. This space is reserved for the engineers in Cohort 4.

The first review in this column could carry your name.

22 live sessions, 6 projects, one showcase.

Enroll now →

Straight from the live sessions

What engineers typed in the cohort chat, unedited.

Messages from past cohorts, posted during and right after the weekend sessions.

HardeepWalia 6:39 PM

@Sid [Team Codebasics] , Brother Your Teaching pattern much better the Speech also in same pattern better pauses ,no low voice in between much clear voice, Thanks for effort

1 reply

Deepak 07:28 PM

you are rockstar

Zia 07:03 PM

truly an interesting ..deep-agents with deep insights :)

Venu Gopal 07:04 PM

Thank you @Dishant Parikh

Siddhartha Banerjee 4:06 PM

It matters a lot in Enterprise scenarios. I am so glad @Dhaval Patel (Team Codebasics) and @Sid [Team Codebasics] you guys are also focussing a session on personal bradning

🙌 1

Pranathi 05:43 PM

Inspiring and motivating session :)

Sergio Yunes 7:02 PM

Thank you very much, Mr. Sid and Mr. Dhaval.

Asish Kumar Mishra 7:03 PM

thank you @Sid [Team Codebasics] & @Dhaval Patel (Team Codebasics) for the session. nice anime @Mochitha [Codebasics Team]. thank you

Kriti K 07:25 PM

thanks to the admin team as well.. the content is nicely organised between discord, recordings and relevant materials to read ..

❤️ 1

g chaitanya 06:30 PM

This session is the Gold Mine!! 💯

❤️ 6

Mulu 07:13 PM

Thank you - We appreciate how you ALL came together as a team to help us understand the concepts today.

✨ 1

Chaithanya Krishna GR 06:22 PM

its always confusing on where to start this helped us to channel the energy. Thanks team

❤️ 2

Zia 06:23 PM

amazing session.. thank you @Hem [Team Codebasics] and @Sid [Team Codebasics]

Suganiya S 7:08 PM

Nice Session ..But still I am yet to get the spark trigger..Will soon hit it..Slowly my confident is improving on the way to build AI use case in my current company ..Thanks ❤️👍

Lavanya 06:31 PM

Thanks to the codebasics team. @Sid [Team Codebasics] , @Dhaval Patel (Team Codebasics) . You are a great inspiration!

Radhika Dhandapani 07:15 PM

Thank you @Sid [Team Codebasics] for all your patience and efforts to bring complex concepts in more understandable way. Really appreciate the amount work you putting behind the screens. Thank a lot.

❤️ 2👍 1👏 1💯 1

Abhishek Timmaraju 07:02 PM

It was indeed a lot, deep and interesting session. Thanks Sid & Dishant

Before you decide

Three ways to learn this. Only one makes you ship.

	Self-study	Generic AI course	This cohort
Live, with instructors	No	Rarely	22 live sessions
What you build	Tutorials	One or two demos	6 graded projects and a showcase
The 2026 stack (LangGraph, MCP, AgentCore)	On your own	Often dated	Updated for Cohort 4
A production bar (evals, latency, cost)	Rarely	Sometimes	Every project
Help between sessions	None	A forum	Discord, all week
Investment	Your time	Usually far higher	US$660 inner circle

Questions

The things engineers ask us.

What do I do on weekdays, after the weekend live sessions?

These are long, technical sessions - what you learn, you need to revise, apply, and build on your own during the week. That takes time, but you're not doing it alone: the Discord stays active all week for doubt-clearing.

Do I also get the AI Engineering & Data Science Bootcamp 4.0?

Yes. Every enrollment includes full access to the AI Engineering and Data Science Bootcamp 4.0, worth ₹18,000, at no extra cost.

When are the live sessions?

Saturdays and Sundays, 5 PM - 8 PM IST. Sessions are fully live and interactive, with a strong focus on hands-on practice, real-time problem solving, and Q&A. Recordings are available for revision.

When does Cohort 4 officially launch?

On Sunday, 25th October 2026.

What if I miss a live session?

All live sessions are recorded and uploaded within 4-6 hours. You can watch them at your own pace.

What happens after the 12 weekends? Do I lose access?

No. You keep access to all recordings for 1 year from your enrollment date.

Do I need ML experience?

No. If you have at least 2 years of software engineering or coding experience, you're ready. We cover AI from an engineering lens.

I'm a fresher or have less than 2 years of experience. Can I join?

We strongly advise against it. This bootcamp moves fast and assumes you can already write code confidently. If you're early in your career, start with the AI Engineering & Data Science Bootcamp first. Build the foundation, then come back.

Who is this bootcamp designed for?

Software engineers moving into AI roles, developers adding AI to existing products, and tech leads or architects designing AI systems. If you write code for a living and want to build real AI systems, this is for you.

How do I get help if I'm stuck?

You can ask questions anytime on Discord. During live sessions, instructors solve problems in real time. You're never stuck alone.

Is there job assistance?

The AI Engineering Bootcamp focuses on building your skills, portfolio, and online presence. The AI Engineering & DS Bootcamp (included with your enrollment) has dedicated job assistance: resume builder, LinkedIn optimization, portfolio website, and mock interviews.

I already own the AI Engineering & DS Bootcamp 4.0. What do I pay?

The amount you paid for the AI Engineering & DS Bootcamp 4.0 is fully adjusted and deducted from your AI Engineering Bootcamp price.

Can I purchase only the AI Engineering Bootcamp without the AI Engineering & DS Bootcamp 4.0?

We don't offer a standalone version. The AI Engineering & DS Bootcamp 4.0 is intentionally bundled as a foundational reference throughout the program. If you already own it, you're eligible for a reduced price and your earlier purchase is not wasted.

I used a subsidy (my existing AI Engineering & DS Bootcamp 4.0 or individual course purchase). Can I refund my original purchase after enrolling?

No. Once your existing purchase is applied as a subsidy to reduce your price, that original purchase becomes non-refundable.

I used a subsidy and now want to refund the AI Engineering Bootcamp itself. What happens?

You get back the amount you actually paid for the AI Engineering Bootcamp. Your original purchase stays intact and you keep access to it.

What system configuration do I need?

Recommended: • OS: Windows 11 • Processor: Intel Core i7 (10th Gen+) or AMD Ryzen 7 (4th Gen+). An i5 works if you're not focused on local model training. • RAM: 8GB minimum, 16GB recommended • Storage: 512GB SSD strongly recommended • GPU: NVIDIA GTX 1660 or higher for deep learning and GPU-accelerated tasks This covers all bootcamp work comfortably. You'd only need stronger hardware if you plan to fine-tune small LLMs locally.

What's the refund policy for Cohort 4?

Full refund if requested by 1st November 2026. After that date, no refunds are available. Check the complete policy: https://codebasics.io/refund-policy

What is the difference between AI Engineering and Data Science Bootcamp 4.0 and the AI Engineering Bootcamp?

AI Engineering & DS Bootcamp 4.0 is self-paced and beginner-friendly, covering AI and Data Science fundamentals from scratch. AI Engineering Bootcamp is a fast-paced, live cohort for experienced software engineers with 2+ years of experience, that assumes strong programming fundamentals and focuses on AI architecture, system thinking, production engineering & shipping real-world AI systems.

What is the inner circle price, and when does it close?

US$660 until Tuesday, 20th October 2026. After that it's US$840 until enrollment closes on Sunday, 1st November 2026. EMI available.

Cohort 4 · Starts Sunday, 25^th October

Twelve weekends from now, one of those two lines is yours.

22live sessions
6projects
1final showcase

Inner circle price

US$660

Closes Tuesday, 20^th October. Then US$840.

Enroll now →

Full refund until Sunday, 1^st November

US$660 inner circle Closes Tue, 20^th Oct, then US$840

Enroll now →