Home → AI & Data Science Full Stack → Gen AI & Agentic AI Full Stack → SAP AI with BTP Full Stack Faculty Why Us Blog Careers Contact Book Weekend Demo

Trending Program

Gen AI & Agentic AI Full Stack

A focused 120-hour program on the fastest-moving part of the AI stack — large language models, RAG, LangGraph, CrewAI, Model Context Protocol (MCP) and production multi-agent systems. Choose a 3-month weekday (Mon–Fri) batch or a 5-month weekend (Sat–Sun) batch.

Book Free Weekend Demo View Full Curriculum

Developer building autonomous AI agents

Team building generative AI applications together

Indian AI product team reviewing an agentic workflow presentation

Team presenting a generative AI agent demo

Who this course is for

Developers & freshers

Programmers with basic Python who want to specialise directly in the highest-demand GenAI/agentic skillset without a full 6-month detour.

Working professionals

Data scientists, software & QA engineers upskilling to build and ship LLM-powered products and autonomous agents at work.

Full curriculum — 120 hours across 11 modules

Every module blends theory with hands-on labs. Click a module to see topic-level detail.

Evolution of LLMs & the GPT/Claude/Gemini landscape
Transformer architecture explained
Tokenization & embeddings
In-context learning & few-shot prompting
Prompt engineering techniques
Context windows & model limitations

OpenAI API (GPT models)
Anthropic Claude API
Google Gemini API
Open-source models via Hugging Face
Running local models with Ollama
Streaming responses & function/tool calling
Cost, latency & rate-limit management

Chunking strategies for documents
Embedding models & similarity search
Vector databases: Pinecone, Chroma, Weaviate, FAISS
Building retrieval pipelines
Hybrid search (keyword + semantic)
Re-ranking & query rewriting
RAG evaluation & guardrails
Working with structured + unstructured sources

Chains, prompts & output parsers
Memory management (short & long-term)
Tool & document loaders
LlamaIndex for data indexing & retrieval
Building a document Q&A assistant
Debugging & tracing with LangSmith

Agent design patterns: ReAct, Reflection, Plan-and-Execute
LangGraph: state machines, nodes & conditional edges
CrewAI: role-based multi-agent collaboration
AutoGen: conversational multi-agent systems
OpenAI Agents SDK & Assistants API
Semantic Kernel overview
Choosing the right framework for the job

MCP architecture: servers, clients & hosts
Building an MCP server to expose tools/APIs
Connecting agents to databases, SaaS apps & browsers via MCP
Secure tool access & permissioning
Integrating MCP with LangGraph/CrewAI agents
Hands-on: building a custom MCP tool integration

Agent-to-agent communication patterns
Task planning & delegation
Short-term & long-term memory systems
Human-in-the-loop workflows
Error handling & fallback strategies
Building a multi-agent business assistant

When to fine-tune vs. prompt vs. RAG
LoRA & QLoRA fundamentals
Dataset preparation for fine-tuning
Prompt optimization techniques
Cost & latency optimization strategies

LLM evaluation frameworks & metrics
Tracing & observability with LangSmith
Guardrails & content moderation
Hallucination detection & mitigation
Responsible AI & data privacy considerations

Building backends with FastAPI
Front-ends with Streamlit / Gradio
Docker containerization for AI apps
Cloud deployment basics
CI/CD for AI applications

Capstone 1: Autonomous multi-agent business assistant
Capstone 2: RAG-based enterprise chatbot with MCP tool integration
Resume & portfolio review
Mock technical interviews

Tools & technologies you'll master

Career outcomes

Gen AI EngineerAI Agent DeveloperPrompt EngineerLLM Application DeveloperConversational AI Engineer

Certificate of Completion

A credential you can show employers

Every graduate receives a verifiable AiTeky certificate of completion. Here's a sample of what yours will look like.

Sample AiTeky certificate of completion for the Gen AI & Agentic AI Full Stack course