Agentic AI Bootcamp: 50 Hours Live Code Along - Deploy a Real AI App
Nitin Monga
AI Tech Founder
10 Weeks to Deployed AI product you can show clients, Employers or Investors
4 cohorts · 5.0★ rating · alumni at ICICI Bank, National Bank of Kuwait, JP Morgan Chase, Bluemarlin, LiveRamp, Oracle, Ogilvy, Meta
You've watched the tutorials. You still can't ship an agent.
That's not a knowledge problem — it's a reps problem. Tutorials show you code. They never make you write it, debug it, and deploy it with someone who's done it before sitting next to you.
This bootcamp fixes that. Every single session is live. Every topic is code-along. You open your editor, I open mine, & we build multi-agent systems with LangChain, LangGraph & MCP line by line — 30 live sessions, the most live instruction of any AI bootcamp on Maven.
And it ends with something no tutorial gives you: a production agentic AI app running on AWS under your own custom domain. Not a notebook. Not a localhost demo. A live URL you can put in front of employers, clients, and investors.
You graduate with:
A deployed multi-agent AI application — live on the cloud, on YOUR domain
Multi-agent orchestration: LangChain + LangGraph + MCP + Agentic RAG & Vector Databases
Evaluation & monitoring so your agent is trustworthy, not just impressive
More Details -https://tinyurl.com/agentic-ai-future
What you’ll learn
Master LangChain, LangGraph, Agentic RAG, MCP - create the intelligent, autonomous AI workflows & AI Agents like - https://www.aitripy.com
Architect an autonomous Multi AI Agents that collaborate & interacts with tools, APIs, and users end-to-end.
Containerize with Docker and deploy on AWS ECS behind your own custom domain
Leave with a live URL — the portfolio piece employers and clients actually click
Build stateful agent workflows with LangGraph: routing, cycles, human-in-the-loop
Design multi-agent patterns — planner/executor, supervisor, agent-to-agent handoffs
Connect agents to tools and data through MCP servers you build yourself
Ingest data with loaders, splitters, and vector DBs ( ChromaDB, Pinecone/ Qdrant)
Go beyond static RAG: retrieval planning, reranking, and context engineering
Evaluate retrieval quality so you can prove your system works — not just hope
Trace every agent step with observability tooling from week one
Build eval loops — golden datasets and LLM-as-judge — to catch failures before users do
Add guardrails against prompt injection, PII leaks, and runaway tool calls
Serve your agent with FastAPI and ship a working UI with Streamlit
Use Claude Code and Cursor as pair programmers to accelerate every build
Wire CI/CD so deployments are repeatable, not heroic
Present your capstone at Demo Day — ready for LinkedIn, interviews, clients, or investors
Learn to scope and pitch agent projects the way a consultant or founder would
Walk into AI engineering interviews able to explain and defend every decision in your system
Who this course is for
Developers moving into AI engineering
You can code, but agents still feel like magic. In 10+ weeks of live building you'll close the gap
Entrepreneurs & Founders
Build your MVP or first client-ready agent during the course. Launch AI-powered products, or automate workflows
Product Managers
Looking to lead AI-driven product development, understand agent architecture & collaborate effectively with technical team
What's included
Live sessions
Learn directly from Nitin Monga in a real-time, interactive format.
50+ Hours of Live - Code Along Sessions
50+ hours of live, code-along sessions - the most live AI Bootcamp on Maven. You code in every class. No passive watching. You build real agentic systems alongside me in real time. You write every line, debug every error, and ship working agents. By the end, you won't just understand agents; you'll have built them.
Deploy Your Own Agentic AI App on a Custom Domain
Build and deploy a production-ready Agentic AI application using LangChain and LangGraph—hosted on AWS under your own custom domain (just like www.aitripy.com). By the end of this course, you won’t just learn—you’ll launch. Showcase your AI expertise with a live, interactive application that employers, clients, and peers can access anytime.
Community of Peers
Stay accountable and share insights with like-minded professionals.
Certificate of Completion
Share your new skills with your employer or on LinkedIn.
Exclusive Community Access
Collaborate, learn, and grow with a vibrant network of AI builders, innovators, and professionals. Stay engaged through discussions, project
The Agent Builder's Library
100+ pages of guides, 20+ architecture diagrams, every repo we build in class, plus my personal templates and decks. Lifetime access, updated each cohort.
All Sessions are Recorded
If you’d like to revisit or catch up on any missed sessions, you can go through the recordings—all sessions are recorded and available on Maven.
Lifetime Access
Go back to course content and recordings whenever you need to.
Weekly office hours
Bring your bugs; we fix them together. Clarify all your doubts.
Maven Guarantee
Your purchase is backed by the Maven Guarantee.
Course syllabus
28 live sessions • 6 lessons
Week 1
Building AI Agents - Orientation
Aug
1
Agentic AI BootCamp - Orientation & Infrastructure Readiness
Week 2
Introduction to Large Language Models (LLMs)
Aug
3
Hands-On ->Integrating LangChain with Large Language Models - OpenAI, Claude, Gemini
Aug
5
LangSmith - Tracing AI Applications, Monitoring Agents & Cost, Latency & Token Analytics
Aug
8
LangChain & ChatPrompTemplate & Messages & Introduction to Structured Output
Schedule
Live sessions
4-5 hrs / week
For US Time Click: https://bit.ly/live-us-time
Sat, Aug 1
3:00 PM—5:00 PM (UTC)
Mon, Aug 3
12:30 AM—2:30 AM (UTC)
Wed, Aug 5
1:30 AM—2:30 AM (UTC)
Projects
1-2 hrs / week
Async content
1-1 hr / week
Frequently asked questions
Projects you will build
01 · YouTube RAG Chatbot — question any video's transcript
02 · RAG with Pinecone — production vector search with metadata filtering
03 · LangGraph Conversational RAG — persistent memory across sessions
04 · LangGraph SQL Agent — plain English → SQL, self-correcting queries
Then it gets seriously agentic.
05 · Multi-Agent Research Assistant — a supervisor delegates to specialists and assembles a sourced brief
06 · Agentic RAG, Hybrid Retrieval — a router picks vector DB or SQL, then synthesizes both
07 · Social Media Multi-Agent System — research → write → review pipeline with a human-approval gate
08 · ReAct Agent with Full Memory — reason, act, observe, remember
09 · Your Own MCP Server — expose custom tools any agent (including Claude) can use
Every repo is yours forever. If it's in your portfolio, you wrote it.
Capstone Project
A production multi-agent system, live on your own domain. Everything converges into one build — a full multi-agent application where specialized agents research, reason, retrieve, and collaborate to complete real tasks end to end. You'll architect it with LangGraph orchestration, MCP tools, a FastAPI backend, a Streamlit front-end, Docker, and AWS. Then we deploy it together to your own custom domain and you present it at Demo Day.
Employers don't click GitHub links. They click yours AI App deployed
More Alumni Reviews
Maven for Teams
Reimbursement
Get your company to pay
Everything L&D needs: email template, receipts, and certificate of completion.
Get reimbursedPrivate cohort
Run a cohort for your org
A dedicated cohort with a custom schedule and curriculum, tailored to your team.
Book a private cohort$1,299
USD
6 days left to apply