Build a Chatbot with Python, RAG and OpenAI training course

 

 

 

 

 

Module 3 — Embeddings and Semantic SearchGoal: Understand how machines “understand meaning” in text.

Topics covered

    • What embeddings are

    • Vector representations of text

    • Cosine similarity

    • Semantic search vs keyword search

    • Introduction to vector databases

Practical lab

    • Generate embeddings

    • Compare similarity between sentences

    • Implement simple semantic search

Outcome: Attendees understand the foundation of retrieval systems.

Module 4 — Document Indexing for RAGGoal: Prepare documents for retrieval.

Topics covered

    • Why chunking matters

    • Chunk size strategies

    • Metadata storage

    • Building an embedding index

    • Loading text/PDF documents

    • Vector storage concepts

Practical lab

    • Load documents

    • Split into chunks

    • Generate embeddings

    • Store in a vector database (e.g. Chroma)

Outcome: Attendees can build a searchable knowledge index.

Module 5 — Retrieval-Augmented Generation (RAG)Goal: Build a chatbot that answers questions using external knowledge.

Topics covered

    • The RAG pipeline

    • Query embedding

    • Retrieving relevant chunks

    • Prompt grounding

    • Hallucination reduction

    • Context injection patterns

Practical lab - Build a RAG chatbot that:

    • accepts a question

    • retrieves relevant document chunks

    • generates a grounded answer

Outcome: Attendees build a real knowledge-based chatbot.

Module 6 — Conversation Memory and ImprovementsGoal: Make the chatbot feel more natural and reliable.

Topics covered

    • Short-term vs long-term memory

    • Conversation summarization

    • Token limits and context management

    • Prompt templates

    • Guardrails and instruction tuning

Practical lab

    • Add memory summarization

    • Improve chatbot reliability

    • Implement structured prompts

Outcome: Attendees can extend chatbot capabilities beyond basic RAG.

Module 7 — Deploying a Chatbot (Optional)Goal: Turn the chatbot into an application.

Topics covered

    • FastAPI chatbot endpoint

    • Simple web UI (Streamlit or similar)

    • Session handling

    • Environment configuration

    • Deployment basics

Practical lab

    • Build a chatbot API

    • Connect a minimal UI

Outcome: Attendees deploy a working chatbot service.