Skip to content
NextiaNextia Learning

Portfolio Capstone: Build, Evaluate and Ship an AI Product

Build, evaluate and ship your own AI product, with the evidence to show it works. You watch a complete document assistant for two shops being built step by step, and you build your own beside it, one lesson at a time.

Start course See the curriculum

Read every lesson as a guest, or sign in to enroll and track your progress.

What you'll learn

By the end of this course, you will be able to:

  • Scope an achievable AI product.
  • Integrate data, AI, API and user interface.
  • Demonstrate evaluation, access control and failure handling.
  • Present a reproducible deployed project with honest limitations.

About this course

Grace runs the help desk at Larkfield. Ines owns Bramble Books. Both shops have policy documents that their staff search every day, and both want an assistant that answers from their own documents, cites them, and says clearly when the documents do not answer. Neither shop may ever see the other shop’s documents.

In this course, Sam builds that assistant as their portfolio project, and you build your own product beside it. Every lesson has two parts. First you see Sam’s worked capstone: the real code, the real outputs and the real mistakes. Then you do the same step for your own project and keep the evidence.

You start small: one user, one decision, a narrow set of documents and acceptance criteria you can test. You reserve the held-out questions before you tune anything. Then you design the data, the API and the database, and you make one complete journey work: upload a document, watch it get indexed, ask a question and get an answer with citations.

Next you make it safe and reliable, you measure it on the held-out questions, and you improve one thing with evidence. Finally you ship a reproducible demonstration and present it honestly, with what works, what fails and what you would do next. You can follow the standard track with the supplied data, or bring your own idea.

Before you start

You completed these courses, or you have the same skills: Building Reliable Applications with LLM APIs; RAG: Building AI That Uses Your Data; Evaluating and Testing AI Systems; Git and Team Development Essentials; Build and Deploy Your First API; Docker and Cloud Essentials for AI Developers; CI/CD, Testing and Safe Releases; Databases and Storage for AI Applications; Authentication, Permissions and Multi-Tenant Applications; and AI Security, Privacy and Responsible Design. The frontend course, Frontend Architecture and Performance for AI Apps, is planned but not published yet, so this course uses one plain HTML page and teaches what it needs. You do not need an account or money. Everything runs on your computer with Docker, and the AI answers are recorded by the course team; live model calls and a cloud deployment are optional.

Courses to take first: Building Reliable Applications with LLM APIs, RAG: Building AI That Uses Your Data, Evaluating and Testing AI Systems, Git and Team Development Essentials, Build and Deploy Your First API, Docker and Cloud Essentials for AI Developers, CI/CD, Testing and Safe Releases, Databases and Storage for AI Applications, Frontend Architecture and Performance for AI Apps, Authentication, Permissions and Multi-Tenant Applications, AI Security, Privacy and Responsible Design.

Who this is for

People who finished the engineering path, or have the same skills, and now want one complete, honest project for their portfolio. You can build an API with a database, a RAG assistant, a multi-tenant sign-in and a test pipeline, as in the earlier courses. Now you put them together, measure the result and present it.

How we made this course

Experienced AI practitioners designed this course. AI helped us draft the content; people chose the teaching approach and the examples, and reviewed every lesson. The outputs that you see are the actual results of running the code. How we make our courses

Inside the course

Module 3Build one complete user journey

3 lessons

Make one user journey work from end to end. Build the backend with a mock AI adapter first, then a plain page for upload, job status and questions, then real retrieval with citations and traces.

  1. Backend foundation120 min
  2. Frontend foundation100 min
  3. Integrate retrieval120 min

Module 4Add reliability and access control

3 lessons

Make it safe and reliable. Enforce sign-in, memberships and row-level security, handle timeouts, bad files and repeated uploads, and let users see the evidence, get an honest "no answer" and delete what they uploaded.

  1. Permission enforcement110 min
  2. Failure handling100 min
  3. User review and cleanup90 min

Module 5Evaluate and improve

3 lessons

Measure it honestly. Run the held-out questions and the isolation checks, sort the errors by cause, make one improvement backed by evidence, and report the trade-offs, costs and regressions.

  1. Run held-out tests100 min
  2. Inspect error categories90 min
  3. Document trade-offs80 min

Module 6Deploy and operate

3 lessons

Ship it so others can run it. Build a tested image with a pipeline, release a private demonstration with secrets, health checks and logs, and prove rollback, cleanup and usage limits.

  1. Build pipeline90 min
  2. Release the demo90 min
  3. Recovery and cost control80 min

Module 7Prepare the portfolio presentation

3 lessons

Present your work. Write the handoff documentation, prepare a demo with a failure case and a trade-off, and reflect on what you learned and what you would build next.

  1. Repository handoff80 min
  2. Demo narrative70 min
  3. Reflection and extensions60 min

Final module · choose at least oneCase studies

2 case studies

Two case studies. Apply the same method to a public document collection with a clear licence, and review a weak capstone against the rubric. Complete at least one for the certificate.

Complete at least one case study for the certificate. The others are optional. What each one covers

  1. Capstone on a public document collection240 min
  2. Review a weak capstone180 min

After the case studiesFinal assignment

Use every module in one practice project. You collect evidence, and check it against the course outcomes.

  1. Final assignment180 min

Case studies

Realistic projects from start to finish, on real data, with every output explained. Each one makes a portfolio piece. Complete at least one for the certificate; the others are optional.

  1. Capstone on a public document collection4 hr

    Apply the same method to an openly licensed document collection: scope, data card, held-out questions, isolation and an evaluation report.

  2. Review a weak capstone3 hr

    Mark a constructed weak capstone against the rubric, find the missing evidence, and write the feedback that would make it pass.

Reviews

Loading the reviews…

    Took this course? Sign in to review it. Learners can write a review after they complete 25% of the lessons.

    Share this course