PSI AI Academy · Tier 2

AI-Enabled Software Development

For developers who want AI in every stage of their work. Over twelve sessions you plan, design, build, test, review and ship real software with AI, and keep your own judgment in charge of the result.

Length: 24 hours
Sessions: 12, most of them 2 hours
Tools: Claude Code, OpenCode, Copilot or Cursor
Finale: 4-hour team capstone
Level: Working developers
Certificate: Certified AI-Powered Software Engineer

Curriculum

Twelve sessions, from requirements to deployment

The course follows a project's life: plan it, build it, ship it, then prove it in a team build.

Plan · 4 hrs
Build · 8 hrs
Ship · 6 hrs
Finish · 6 hrs

Plan01–02

01
AI-Augmented Requirements and Planning
Extract requirements, refine user stories and generate technical documentation.
2 hrs
02
AI-Assisted System Design and Architecture
Prototype user experiences, evaluate architecture and write API specifications.
2 hrs

Build03–06

03
Foundations of Coding with AI
Work in Claude Code, OpenCode, Copilot or Cursor, and learn to prompt them well.
2 hrs
04
Design Thinking with AI
Use AI for ideation and rapid prototyping to reach a working solution.
2 hrs
05
Advanced Code Generation and Refactoring
Generate complex modules, improve performance and refactor legacy code.
2 hrs
06
AI-Assisted Debugging and Code Review
Diagnose errors, find vulnerabilities and review code with AI.
2 hrs

Ship07–09

07
AI-Powered Testing
Generate unit, integration and end-to-end tests, then read the coverage.
2 hrs
08
Documentation and Knowledge Management
Write API documentation, READMEs, guides and runbooks with AI.
2 hrs
09
AI in CI/CD and Deployment
Automate pipelines, deployment workflows and monitoring.
2 hrs

Finish10–12

10
Responsible AI in Software Development
Security, privacy, licensing and ethics, with a human signing off.
1 hr
11
Capstone Project (Hackathon)
Build a working application in teams, using AI at every stage. How it runs
4 hrs
12
Wrap-Up and Professional Development
Show your project, measure what AI changed, and plan what to learn next.
1 hr

Who it is for

Developers ready to work with AI every day

Software developers
Junior to mid-level engineers bringing AI into real projects.
CS and IT students
Upper-year and graduate students preparing for engineering work.
QA and DevOps
Technical staff who want AI help with testing and delivery.
Team leads
Leads judging how AI can change their team's workflow and quality.

Capstone · Session 11

A four-hour team build

Small teams build a working application in one sitting, with AI at every step.

1Define the requirements
2Design the architecture
3Build and refine with AI
4Test, secure and deploy
5Show the application

FAQ

Questions

Is this for beginners?

No. You should already program and know how a development project runs.

Which AI tools do we use?

Claude Code, OpenCode, Copilot or Cursor, alongside the testing tools and CI/CD you would use on a real project.

Is it hands-on?

Yes. Most sessions have exercises, and the course ends with the four-hour capstone.

How long is each session?

Most sessions are two hours. Responsible AI and the wrap-up are one hour each, and the capstone is four hours, for 24 hours in all.

Do I work alone or in a team?

Both. You work through the session exercises yourself, then build the capstone with a small team.

What will I have at the end?

A working application your team built and presented in the capstone, and a plan for what to learn next.

Does it cover using AI responsibly?

Yes. Session 10 covers security, privacy, licensing and ethics, and every session keeps a human signing off on what AI produces.

When does the next run start?

Do I get a certificate?

Yes. Pass the assessment and the capstone to become a Certified AI-Powered Software Engineer, issued by Predictive Systems, Inc.

How much does it cost?

Can we run it for our company?