Skip to content
An engineer building an AI chat assistant with cited sources on a large monitor

4 Months · LLM apps that ship

Generative AI & LLM Engineering

Build production Generative AI applications: prompting, embeddings, RAG, agents and fine-tuning, shipped as real services. For developers and professionals moving into AI, in Chennai and online.

Book a Demo See the Roadmap
  • 7Phases
  • 4Months
  • 100%Project Based
Python◆ REST APIs◆ Git◆ Transformers◆ Prompting◆ Structured Output◆ Embeddings◆ pgvector◆ Pinecone◆ RAG◆ LangChain◆ LlamaIndex◆

Phases 1–2

Python for AI Apps

Python for application work: functions, classes, async, calling REST APIs, handling JSON and keeping code in Git.

01
Phase 1

Python for AI Apps

Python for application work: functions, classes, async, calling REST APIs, handling JSON and keeping code in Git.

PythonREST APIsGit
02
Phase 2

LLM Foundations & Prompting

How large language models work: tokens, context windows, transformers in plain terms, prompt patterns, structured JSON output and basic evaluation.

TransformersPromptingStructured Output

Phases 3–4

Embeddings & Vector Search

Turning text into embeddings, chunking documents, storing them in a vector database, hybrid search and reranking for better results.

03
Phase 3

Embeddings & Vector Search

Turning text into embeddings, chunking documents, storing them in a vector database, hybrid search and reranking for better results.

EmbeddingspgvectorPinecone
04
Phase 4

RAG Applications

Retrieval-augmented generation end to end: document ingestion, retrieval pipelines, answers with citations and measuring answer quality.

RAGLangChainLlamaIndex

Phases 5–6

Agents & Tool Use

Function calling, multi-step agents that use tools and APIs, agent graphs, memory, and guardrails that keep them safe.

05
Phase 5

Agents & Tool Use

Function calling, multi-step agents that use tools and APIs, agent graphs, memory, and guardrails that keep them safe.

Tool CallingLangGraphGuardrails
06
Phase 6

Fine-tuning & Open Models

When to fine-tune instead of prompting, LoRA and QLoRA with Hugging Face, and running open models on your own machine.

Hugging FaceLoRAOllama

Phase 7

Ship It: LLMOps

Serving an AI feature as an API with streaming, Docker deployment, cost and latency control, monitoring and evaluation in production. Capstone project.

07
Phase 7

Ship It: LLMOps

Serving an AI feature as an API with streaming, Docker deployment, cost and latency control, monitoring and evaluation in production. Capstone project.

FastAPIDockerLLMOps

Curriculum

Every phase, in order.

7 phases over 4 months. Each one ends with something you've built and can show, not a quiz you've passed.

01
Phase 1

Python for AI Apps

Python for application work: functions, classes, async, calling REST APIs, handling JSON and keeping code in Git.

PythonREST APIsGit
02
Phase 2

LLM Foundations & Prompting

How large language models work: tokens, context windows, transformers in plain terms, prompt patterns, structured JSON output and basic evaluation.

TransformersPromptingStructured Output
03
Phase 3

Embeddings & Vector Search

Turning text into embeddings, chunking documents, storing them in a vector database, hybrid search and reranking for better results.

EmbeddingspgvectorPinecone
04
Phase 4

RAG Applications

Retrieval-augmented generation end to end: document ingestion, retrieval pipelines, answers with citations and measuring answer quality.

RAGLangChainLlamaIndex
05
Phase 5

Agents & Tool Use

Function calling, multi-step agents that use tools and APIs, agent graphs, memory, and guardrails that keep them safe.

Tool CallingLangGraphGuardrails
06
Phase 6

Fine-tuning & Open Models

When to fine-tune instead of prompting, LoRA and QLoRA with Hugging Face, and running open models on your own machine.

Hugging FaceLoRAOllama
07
Phase 7

Ship It: LLMOps

Serving an AI feature as an API with streaming, Docker deployment, cost and latency control, monitoring and evaluation in production. Capstone project.

FastAPIDockerLLMOps

Career Deliverables

What you walk out with.

Opportunities waiting for you — and the evidence to go after them.

01

Deployed projects

Live URLs and a public repo, not screenshots.

02

Verifiable certificate

Checkable by any employer at any time.

03

Interview practice

Technical and HR rounds with written feedback.

04

Placement support

Referrals and applications until you land the role.

Start the
GenAI track.

Enrol Now All Courses