Students bring their own notes, PDFs, and resumes into these projects — personal stakes are part of the curriculum design, not an accident.
Project 01
AI Study Assistant
3 hours · Sessions 6–7
Your first shipped AI product — a personal tutor you actually use.
A conversational assistant that answers subject questions, explains concepts in plain language, and summarizes your own notes. You pick your own subject domain, so the assistant is immediately useful for your actual coursework.
What it does
- Question answering & concept explanation
- Note summarization
- MCQ generation for self-testing
- Study-plan generation
Skills you practice
LLM API calls from PythonFastAPI backend with streaming responsesPrompt engineering for structured outputs
Understand the LLM behind it →
Project 02
Chat With Your Documents
4 hours · Sessions 9–10
A ChatGPT that has actually read your PDFs.
A full Retrieval-Augmented Generation (RAG) pipeline: upload your own textbooks, notes, or research papers, and ask questions that get answered with citations back to the source text — not a hallucinated guess.
What it does
- PDF upload & parsing
- Chunking + embeddings
- Vector search with Chroma
- RAG-based answers with source references
Skills you practice
Vector databases & semantic searchEmbeddings & chunking strategyRetrieval + context injection
Read: What is RAG? →
Project 03
AI Career Assistant
5 hours · Sessions 12–14 · Capstone
The capstone: every skill from the course, combined into one product.
An AI agent that analyzes your resume, compares it against a real job description, detects skill gaps, generates a personalized learning plan, and runs a mock interview with answer evaluation — presented live on Demo Day.
What it does
- Resume & job-description analysis
- Skill-gap detection
- Personalized learning-plan generation
- Mock interview + answer evaluation
Skills you practice
Tool calling & agent loopsRAG integrationExternal API orchestrationEnd-to-end product thinking
Read: AI Agents Explained →