// Live Cohort 01
Now enrolling
The right way to become a GenAI engineer.
GenAI & Agents in Backend Systems
// What you build
3 projects · shipped by the end of the cohort
Python Microservices and Async workers
A production-grade repo you ship and defend in review.
01
Enterprise RAG Chatbot
Retrieval over real company documents, with evals that show it works.
02
Agentic HR Application
An agent that calls tools and takes actions, with guardrails and logs.
03
Reserve your seat for ₹5,000. The balance is due before the first session.
Download Brochure
TAUGHT LIVE BY WORKING ENGINEERS
Lead instructor
Venu Madhav
Co-Founder, AltGAN · IIT Delhi
SK
Sai Kumar · Cisco
5+ years experience
AP
Abhishek PG · HCLTech
8+ years experience
Live sessions
Three sessions a week
Format
Build on a real repo
// The curriculum
What's actually inside each project.
// THE SCHEDULE
Eight weeks, three production systems.
Weeks 1 to 2 · Core foundation
Fundamentals coverage
Python/Git workflow/Coding standards/Clean code/Code review
Alongside project 01Taught against real code
Project 01 · Weeks 1 to 2Backend Gym Management Application
Python microservices/Async batch jobs/Auth and RBAC/Payment reconciliation
Week 1Fundamentals against real code, then FastAPI services with clean boundaries, config, health checks and Postgres migrations
Week 2Celery jobs, queues, retries and idempotency, auth and RBAC, payment reconciliation, integration tests and getting the service deployed
You leave with
A running multi service platform and the retry story to explain it.
SK
Owns and reviewsSai Kumar · Cisco
Project 02 · Weeks 3 to 5GenAI Enterprise Chatbot
RAG over enterprise docs/Retrieval evaluation/LLM as judge/Guardrails
Week 3Ingest and chunk a heterogeneous corpus, embeddings and the first answer that holds up
Week 4Hybrid search, re ranking, grounded answers with citations and knowing when to refuse
Week 5Retrieval evaluation, LLM as judge scoring, guardrails, then latency and cost in production
You leave with
An eval suite that shows why the answers hold up.
VM
Owns and reviewsVenu Madhav · AltGAN
Project 03 · Weeks 6 to 8Agents Agentic HR Application
Multi agent orchestration/Tool calling/Human in the loop/Audit trails
Week 6Multi agent orchestration, handoffs and tool calling against real services
Week 7Tool design, retries and recovering when a downstream service fails, with audit trails behind it
Week 8Human in the loop approval gates, evaluation, cost caps and demo day, defending what you built
You leave with
An agent you would let near a real HR inbox.
AP
Owns and reviewsAbhishek PG · HCLTech
// Who's teaching
Real working engineers.
Not trainers. Engineers shipping this kind of work right now.
Venu Madhav
Co-Founder, AltGAN · IIT Delhi
Cohort lead, owns the Enterprise RAG project and its code reviews
Built the AltGAN learning platform. Runs the 1:1 sessions and sorts out any technical issue you hit.
Sai Kumar · Cisco
Leads and owns the Python microservices project and its code reviews
Building backend systems at Cisco, proficient in using coding agents.
Abhishek PG · HCLTech
Owns the Agentic HR project and its code reviews
Builds multi agent automation for enterprise clients at HCLTech.
// How we teach
Same repo. Study it, build it, ship it.
Study · Self-paced
Learn from a system that already works.
- Work through a real repo, not a toy example built to teach one concept
- See the architecture and the decisions before you touch any code
- Go at your own pace during the week, session by session
- Study the trade-offs someone already had to make and why
Build · Live
Watch it built live, then do it yourself.
- Instructor live-codes on the same repo you studied all week
- Watch a real change get planned, built and debugged, live, not scripted
- Ask questions in the moment, on the same code you’ll work in next
- See how someone who owns this system actually thinks through it
Planned
Written
Tested
Debugged
Your instructors ship this kind of work at their day jobs right now. None of them are trainers who left the industry years ago.
Ship · Real tickets
A ticket on the same repo. Reviewed by the engineer who owns it.
In progress
1
Add retry logic to the payment worker
Retry failed charges with backoff so a flaky provider doesn’t drop payments.
Reliability
4h
Continue task →
In review
1
Expose a health endpoint per service
Add /health to each service so the orchestrator can detect failures.
Observability
3h
PR open, awaiting review
Done
1
Containerize the API gateway
Write a Dockerfile and compose config so the gateway runs the same everywhere.
DevOps
5h
The task is not a repeat of what got live coded. It is new work on the same repo, reviewed and merged by the engineer who owns it.
// Coding agents
Use agents to build. Learn to own the result.
You learn both to use coding agents and to build applications that contain agents.
Before generation
Write a focused spec
Behaviour, files, constraints and acceptance criteria.
During implementation
Guide a bounded change
Give context, inspect the plan, keep changes reviewable.
After generation
Review and verify
Read the diff, run tests, check real behaviour.
In your submission
Show your contribution
Note what the agent helped with and what you changed.
// Beyond the cohort
Support that doesn't stop at week 8.
Individual support on blockers, design choices and PR feedback.
A mock GenAI interview based on your actual projects, with feedback.
An instructor helps you present your cohort projects on your resume.
A certificate for the cohort you completed, three shipped projects behind it.
For top performers
Job interview opportunities
Perform well across your tasks and reviews, and you may be connected to real job interviews, subject to openings.
// Pricing
Serious mentorship, without the bootcamp price tag.
Three shipped projects, live reviews and real engineers, for a fraction of what the market charges.
Price
₹29,999
₹24,999
Significantly less than what similar mentorship costs elsewhere.
3 Shipped Projects
Live Reviews
1:1 Sessions
Mock Interviews
Certificate
Or
// Before you join
Questions people ask first.