How to Build a Responsible AI Chatbot Using Open Technologies 7 Sections35 Lessons5 QuizzesLifetime

Expand all sectionsCollapse all sections

Introduction
2 Items
- 1.1
  The CABot Story
- 1.2
  How to Do This Course
Module 1 - Understanding AI
9 Items
Large Language Models (LLMs) generate language, not truth. Explore tokens, prompts, context windows and hallucinations, then examine why Retrieval-Augmented Generation (RAG) grounds responses in trusted knowledge.
Module 2 - Building the Knowledge Base
9 Items
Understand how to transform a digital resource (e.g. a book) into semantic knowledge through chunking, embeddings and vector search. Learn why CABot uses ChromaDB and metadata rather than retraining models.
Module 3 - Choosing Your Architecture
5 Items
Compare hosted APIs, local models, vector databases and orchestration frameworks. Rather than recommending one stack, this module helps you choose technologies appropriate to your own community.
Module 4 - Building the Chatbot
8 Items
In this module, we follow a learner's question from a browser prompt to asynchronous RAG orchestration. We aim to understand asynchronous architecture and browser-first deployment.
Module 5 - Responsible & Multilingual AI
5 Items
Responsible AI begins with trustworthy knowledge, not simply powerful models. Explore transparency, source attribution, multilingual retrieval, privacy, accessibility and environmentally conscious AI.
Course Conclusion & Next Steps
2 Items
- 7.1
  Lessons learned
- 7.2
  Recap: Practical workflow for building a responsible chatbot

The whole journey

When a student enters a question, several specialised components cooperate:

Frontend: The browser displays the conversation and sends the question.
Flask API: The backend checks that the message and session are valid.
Celery queue: The question becomes a background task so the web server does not freeze while the model works.
Redis: It carries task information and stores recent conversation history.
Embedding model: It converts the question into a vector.
ChromaDB: It finds the closest Climate Academy chunks.
Prompt builder: It combines rules, retrieved passages, and the new questions.
Ollama and Llama: The local LLM generates the answer.
Session memory: The question and answer are added to recent history.
Frontend polling: The browser checks periodically until the result is ready, then displays the answer and sources.

No single component ‘is’ the chatbot. The chatbot is the whole system.

Learning objectives