Capstone Project Notes
Subject Overview
The capstone is not coursework. It is a real product, built end-to-end, that solves a real problem for real users, applying everything from the previous three years — the focus is entirely on building. It must use Python and at least one ML or LLM model, be deployed and publicly accessible (not a localhost demo), include quantitative evaluation with documented results, be built with AI assistance (Cursor, Claude Code or equivalent, with the vibe coding workflow itself part of the deliverable), and reach at least five real users whose feedback drives at least one iteration.
Subject at a glance
8
Units
41
Topics
30–40
Hours
8
Infographics
Unit-wise Syllabus
8 units — click WhatsApp below to get the full notes for each
Unit 1: Discovery & Spec (4–5 hours)
- Pick the problem
- talk to potential users
- write a one-page spec
- sketch the architecture
- 1Pick a real problem
- 2Talk to five or more potential users
- 3Write a one-page spec
Users, goals, success metric
- 4Sketch the architecture
Study guide
Discovery and specification: you pick the problem, talk to potential users, write a one-page spec and sketch the architecture. A well-defined problem is half the project, so spend real time on it.
Unit 2: Data & Model (10–12 hours)
- Collect or generate data
- train/fine-tune a model OR design the LLM pipeline
- run baseline evaluations
- 1Collect or generate data
- 2Choose an approach
Train, fine-tune or LLM pipeline
- 3Build a baseline
- 4Run baseline evaluations
- 5Improve on the weakest cases
Study guide
Data and model: you collect or generate data, train or fine-tune a model or design the LLM pipeline and run baseline evaluations. Always build a simple baseline first so you can prove your approach adds value.
Unit 3: Build the Product (15–18 hours)
- Frontend (Next.js/React)
- backend (FastAPI)
- database
- auth
- using Cursor + Claude Code as your pair programmer
- Frontend
Next.js or React
- Backend
FastAPI
- AI layer
Model, RAG or agent
- Data
Database and auth
Study guide
Build the product: you create the frontend (Next.js or React), backend (FastAPI), database and authentication, working with Cursor and Claude Code as your pair programmer. Keep scope small and ship a working version early.
Unit 4: Deploy & Monitor (5–6 hours)
- Deploy to production
- set up observability and error tracking
- add an evaluation harness to CI
- 1Deploy to production
- 2Set up error tracking
Sentry
- 3Add LLM observability
Langfuse
- 4Add an evaluation harness to CI
Study guide
Deploy and monitor: you deploy to production, set up observability and error tracking and add an evaluation harness to CI. A deployed, monitored app is far more convincing than a local demo.
Unit 5: User Testing & Iteration (4–6 hours)
- Onboard real users
- collect feedback
- ship at least one substantial improvement
- 1. Onboard real users:
- 2. Collect feedback:
- 3. Prioritise: Pick the biggest issue
- 4. Ship an improvement:
Study guide
User testing and iteration: you onboard real users, collect feedback and ship at least one substantial improvement. Real user feedback is the most valuable input to any product.
Unit 6: Demo Day (2–3 hours)
- Write-up
- demo video
- public launch (Product Hunt / X / LinkedIn)
- open-source the relevant parts
- 1Write-up
Problem, approach, results
- 2Demo video
Two to three minutes
- 3Public launch
Product Hunt, X, LinkedIn
- 4Open-source what you can
Study guide
Demo day: you write up the project, record a demo video, launch publicly (Product Hunt, X or LinkedIn) and open-source the relevant parts. Treat the write-up as a job-application document.
Unit 7: Example capstone ideas
- Domain-specific copilot — a coding assistant for a niche framework or a writing assistant for a specific industry
- RAG-powered knowledge product — a searchable expert system for legal docs
- medical papers or university lectures
- vision application — a fine-tuned CV model for retail shelf analysis
- agriculture or accessibility
- AI agent for a real workflow — handling support tickets
- generating marketing campaigns or automating a research task end-to-end
- creative tool — an AI-powered video editor
- music generator or design assistant with a polished UI
Domain copilot
Coding or writing assistant for a niche
RAG knowledge product
Legal, medical or lecture search
Vision application
Retail, agriculture, accessibility
Workflow agent
Support tickets, marketing, research
Creative tool
Video, music or design assistant
Study guide
Example capstone ideas include domain-specific copilots, RAG knowledge products, vision applications, AI agents for real workflows and creative tools. Choose something you can test with real users nearby.
Unit 8: Evaluation criteria
- Does it work — real users can use it without hand-holding
- is the AI integration thoughtful — not just an OpenAI wrapper
- with evidence of evaluation
- prompt engineering or model adaptation
- is the code shippable — tests
- basic CI
- error handling
- observability
- did vibe coding accelerate you — a short write-up on how AI tools helped or hindered the build
- did you learn from real users — documented feedback and at least one iteration based on it
It works
Real users without hand-holding
Thoughtful AI
Evals, prompting or adaptation, not a thin wrapper
Shippable code
Tests, CI, error handling, observability
Vibe coding reflection
How AI tools helped or hindered
User learning
Feedback and at least one iteration
Study guide
The evaluation criteria are that it works for real users, the AI integration is thoughtful and evaluated (not just an API wrapper), the code is shippable with tests and CI, you reflect on how AI tools helped or hindered the build and you learned from real users. Keep evidence of each as you go.
Why Capstone Project Matters
The capstone is your proof of ability: a real product, used by real people, with code, evaluation and a public write-up. It is the centrepiece of your portfolio when applying for AI engineering jobs.
How to Prepare for Capstone Project
- Choose a problem with real users you can reach in the first week.
- Document decisions, experiments and feedback as you go.
- Launch publicly and share the repository and demo.
Capstone Project: Frequently Asked Questions
What is the capstone project?
It is a final project in which you build, deploy and test a real AI product with users, then present it publicly.
What makes a good capstone project?
A clear problem, thoughtful AI integration with evaluation, shippable code, real user feedback and a clear write-up.