Save on courses or Exam now
About Instructors
-
andrewjames- Instructor Rating - Reviews - Students - Courses
No biography available, please complete
🚀 Course Goals
- Brief, steer and review an AI coding agent (Claude Code) with repository maps, CLAUDE.md conventions and small, reversible task slices. Build a secure multi-tenant SaaS on PostgreSQL/Supabase: data model, migrations, sign-up, OAuth, organizations, roles and row-level security. Add AI features people can trust: structured outputs, streaming chat, human-approved actions, and RAG answers with citations and refusal. Connect the backend to the outside world safely: API routes, external REST APIs, verified webhooks, background jobs, retries and idempotency.
📘 About This Course
Ask BluePeak which pages the website relaunch covers, and the answer comes back with a quote and a link to the exact passage it came from. By the end of this course you will have built that feature, and the whole multi-tenant SaaS around it, with Claude Code doing much of the typing and you making the decisions.
This is a production-first course for people who want to build real software with AI without being professional developers. Across 17 modules you build one product, BluePeak AI Workspace, from a blank repository to a deployed release: a team workspace with projects, documents, an AI assistant grounded in your own files, notifications, plans and an admin side.
You learn to brief and steer an AI coding agent, keep a project memory, slice work into small reversible steps and review the diffs it produces. Then you add the layers a real SaaS needs: PostgreSQL data modeling and migrations, sign-up, OAuth, organizations and row-level security, APIs, webhooks and background jobs, LLM features with structured outputs, streaming chat and human approval, file uploads, embeddings and RAG with citations, transactional email, Playwright tests, an OWASP-style security review, CI/CD, deployment, observability, analytics and cost control.
Every lesson is a real session on a real lab: you see the prompt, the code, the command, the output and the result in the browser. Each lesson also includes a controlled failure, diagnosed from the evidence and fixed with the smallest possible change, because that is the skill you will need most on your own projects. All data is synthetic, payments use Stripe test mode only and email is caught in a local sandbox, so you can practise everything safely.
What You'll Learn
Brief, steer and review an AI coding agent (Claude Code) with repository maps, CLAUDE.md conventions and small, reversible task slices.
Build a secure multi-tenant SaaS on PostgreSQL/Supabase: data model, migrations, sign-up, OAuth, organizations, roles and row-level security.
Add AI features people can trust: structured outputs, streaming chat, human-approved actions, and RAG answers with citations and refusal.
Connect the backend to the outside world safely: API routes, external REST APIs, verified webhooks, background jobs, retries and idempotency.
Test, harden and ship: unit, integration and Playwright tests, an OWASP-style review, CI/CD with protected branches, deployment and rollback.
Operate and grow the product: logs and correlation IDs, privacy-aware analytics, performance, cost controls, onboarding and maintenance with AI.
🛠️ Requirements
- No programming experience required: a computer that can run Node.js, Git and a browser, plus curiosity and a few hours a week.
- Access to Claude Code (or a comparable AI coding agent) and a free Supabase or local PostgreSQL setup; everything else is installed on camera.
- Payments use Stripe test mode and email uses a local sandbox: no real customer data, card or paid cloud account is needed to follow along.
👤 Who This Course is For
- Founders, product managers, designers and domain experts who want to build and ship a real AI SaaS without being professional developers.
- No-code and visual-builder users who want to move a prototype into a real codebase they can test, deploy and maintain.
- Junior developers and technical freelancers who want a disciplined, review-first way of shipping production software with AI agents.
53618 Students applied
100 Comments
-
-
John DoeMorbi nec nisi ante. Quisque lacus ligula, iaculis in elit et, interdum semper quam. Fusce in interdum tortor. Ut sollicitudin lectus dolor eget imperdiet libero pulvinar sit amet
-
John Doe
Morbi nec nisi ante. Quisque lacus ligula, iaculis in elit et, interdum semper quam. Fusce in interdum tortor. Ut sollicitudin lectus dolor eget imperdiet libero pulvinar sit amet
-
-
John Doe
Morbi nec nisi ante. Quisque lacus ligula, iaculis in elit et, interdum semper quam. Fusce in interdum tortor. Ut sollicitudin lectus dolor eget imperdiet libero pulvinar sit amet
-
John Doe
Morbi nec nisi ante. Quisque lacus ligula, iaculis in elit et, interdum semper quam. Fusce in interdum tortor. Ut sollicitudin lectus dolor eget imperdiet libero pulvinar sit amet
-
John Doe
Morbi nec nisi ante. Quisque lacus ligula, iaculis in elit et, interdum semper quam. Fusce in interdum tortor. Ut sollicitudin lectus dolor eget imperdiet libero pulvinar sit amet
Morbi nec nisi ante. Quisque lacus ligula, iaculis in elit et, interdum semper quam. Fusce in interdum tortor. Ut sollicitudin lectus dolor eget imperdiet libero pulvinar sit amet