Agentic AI Engineering

Build production multi-agent AI systems — in your domain, from day one,

in 40 live hours.

What You Need

IT experience in development, testing, or support
Basic Python — variables, functions, for loops (we cover gaps in M1)
A laptop with 8GB+ RAM (16GB recommended for local models)
Zero AI or ML background — we start from scratch

Who This Is For

Software developers — backend, fullstack, any stack
Support engineers and DevOps professionals
Tech leads architecting AI adoption for their team
QA and automation engineers
Data engineers and pipeline specialists
Product managers who want to understand what's possible

This Is Not Right For You If

You want deep ML theory or research-level content
You need a self-paced course to do in your own time
You need a university-accredited or NASSCOM certificate

Secure your spot today and transform your career with in-demand skills. Act fast—limited seats available!

Ready to Start Building?

Cohort 1 is limited to 20-30 Participants. Seats are on a first come, first served basis.

Kanak is available to answer your questions before you enroll.

Course Overview

40

Live Hours

13

Sessions

3

Domain Tracks

12

Capstone Projects

Choose Your Domain Track

Unlike most AI courses that use generic examples, this one starts with who you already are. You pick a track in Session 1 and stay on it for the whole course — every lab, every capstone, every exercise.

Healthcare / Education

Build agents that handle patient intake, answer medical Q&A, manage clinical knowledge bases, and guide learners through adaptive study plans.

Best for: Healthcare IT, edtech, clinical informatics

Easy access ensures seamless engagement, fostering convenience and user-friendly interactions for everyone involved.

Banking / Finance

Prioritizing clients, our approach revolves around personalized solutions and tailored experiences, meeting individual needs.

Build agents that advise on financial products, screen loan applications, analyse portfolios, and automate compliance workflows over SEBI and RBI documents.

Best for: BFSI, fintech, data-heavy domains

IT Operations

Upholding honesty and transparency, our commitment to integrity ensures trust and reliability in every interaction.

Build agents that classify and resolve support tickets, auto-generate runbooks, monitor services, and trigger automated remediation workflows.

Best for: QA engineers, support leads, DevOps

Why This Course Works

Your Domain, From Day One

Pick Healthcare, Banking, or IT Operations in Session 1. Every lab, every exercise, and your capstone project stays in that domain. You're building something relevant to your career — not a generic chatbot nobody asked for.

A Working Agent in 60 Minutes

Most AI courses take weeks before you have anything to show. In Session 1 — within the first hour — you have a running agent in your chosen domain. Every session after that adds a new capability to the same system, so you always have something working.

A Portfolio Piece, Not a Certificate

The course ends with Demo Day: you present a deployed app live on the internet, hosted on GitHub, accessible from any phone. AI engineering skills are increasingly valued across IT roles — and a working system you built is far more credible evidence of that capability than a certificate.

Minimal API Costs

We run on Groq's free tier — a generous daily allowance with no credit card needed. For local work, Ollama runs models on your own laptop for free. Estimated API spend for the full 40 hours is around ₹3,000, depending on usage. You focus on building, not managing bills.

Courses Outline

From an early stage start-up’s growth strategies to helping existing businesses, we have done it all!

"This course exists because I kept seeing the same gap — IT professionals who were technically strong, who understood systems and production and real constraints — but had no practical path into agentic AI that started with what they already knew." "Everything in this curriculum was designed backwards from one question: what does an IT professional actually need to build a production-grade AI agent? Not what looks good on a course outline. Not what makes a nice demo. What actually works — in a real domain, with real data, deployed on the internet." "If you show up and do the work, you'll ship something you're proud of. That's the promise."

— Ketan | Course Instructor, Niche Thyself

GitHub Copilot for test code generation

Selenium with AI enhancements

Playwright with AI enhancements

API testing with AI validation

Enquire Now

Courses Outline – Self-Paced Learning

From an early stage start-up’s growth strategies to helping existing businesses, we have done it all!

Git/GitHub

Jenkins

BDD/Cucumber

Enquire Now

What This Course Is

If you work in IT — as a developer, tester, support engineer, or data engineer — and AI is reshaping everything around you with no clear path to get hands on it, this is where you start. Not with slides about AI. With a running agent, in your field, built by you, in the first hour of Session 1.No data science background needed. No ML theory. You bring your IT experience; the course gives you the tools, the structure, and 40 hours of live, hands-on building time. By Demo Day, you have a deployed multi-agent AI application with a public URL, on GitHub, with your name on it.

All our trainers are having minimum 10 years of experience in test automation.

Every session we conduct is a combination of theory and hands-on.

All sessions are recorded which participants can keep with them for life time.

All our trainers are having minimum 10 years of experience in test automation.
Every session we conduct is a combination of theory and hands-on.
All sessions are recorded which participants can keep with them for life time.

OUR VALUABLE

The Tech Stack

Every tool here was chosen because it’s what practitioners actually use. Nothing academic. Nothing that will be outdated before you finish the course.

Language

Python

Agent Framework

LangGraph (core) + CrewAI (multi-agent)

Memory & RAG

ChromaDB + pgvector

Security

LlamaGuard + OWASP LLM Top 10

AI-Assisted Coding

GitHub Copilot + VS Code

LLMs

Groq llama-3.3-70b (free tier) + Ollama (local, free)

Tool Protocol

MCP — Model Context Protocol (97M monthly SDK downloads)

Observability & XAI

LangSmith + Arize Phoenix + explain_decision()

UI & Deployment

Streamlit Community Cloud (public URL)

IT Ops Elective

n8n — self-hosted, free, agentic automation across 400+ services

Course Details

Format

Live online sessions via MS Teams

Duration

40 hours across 13 sessions

Session Length

~3 hours each

Hands-on Time

60% — 24 hours of labs and building

Community

WhatsApp group — cohort access throughout the program

Recordings

Available within 24 hours of every session

Office Hours

Weekly catch-up sessions for participants who fall behind

Batch Size

20–30 participants

Course fee details and batch dates will be shared directly.

Common Question

Let’s Clear Your Doubts

1. Do I need to know Python well?

Basic Python is enough — variables, functions, loops, installing packages. If you can write a for loop and call a REST API, you're ready. We cover the remaining gaps in Module 1 so nobody is left behind from the start.

2. How much will I spend on API costs?

Estimated API spend is around ₹3,000 for the full 40 hours, depending on your usage. Groq's free tier covers a significant portion of this — no credit card required to get started. Ollama handles local model runs at no cost. We walk through the full setup together in Session 1.

3. What if I miss a session?

Recordings go up within 24 hours of each session. Weekly office hours are specifically for catch-up — not just Q&A. The course is structured so each session builds on the last, which means staying current matters; we actively help you do that.

4. Will this get me a job in AI?

We don't promise job placement. We do promise a deployed, GitHub-hosted AI system that shows real engineering capability. AI skills are increasingly relevant across IT roles — having a working system you built and can demo is far more convincing to any hiring manager or internal team than a certificate alone.

5. What laptop do I need?

Any modern laptop with 8GB RAM runs the full course using Groq's cloud API. 16GB RAM is recommended if you also want to run models locally with Ollama. No GPU required at any point.

6. Is there a certificate?

Yes — a course completion certificate. More practically, you'll also have a deployed capstone project on GitHub with a live public URL. In a technical interview, that project will do more work for you than a certificate ever could.

Niche Thyself – Benchmark in software testing education Catalyst for transformative tech careers.