Skip to content
agentGreg EN · PLBook 30 min

Training · AI in software development

GitHub Copilot and AI agents training for development teams

Three days on your code, from Copilot's modes and the autonomous agent to spec-driven development and AI code review. I used this programme to train 20 developers at PKO Bank Polski.

In short

Who it's for
Development teams, tech leads and architects, including teams that already have Copilot licences.
Length
1 to 3 days. The full programme takes 3 days.
Format
On-site in Poland or Norway, or remote, in English or Polish, with exercises on your repository.
Price
From PLN 7,000 net per training day. You order directly from me, with no agency in between, and the fixed price is agreed before we start.

Programme

  1. Day 1. Copilot and the agent in practiceCopilot's working modes on a real task, then the fully autonomous agent. We look at where it speeds things up and where it quietly breaks code, and learn to catch it.
  2. Day 2. Keeping the agent under controlRepository instruction files, custom agents for repetitive tasks, hooks and skills, MCP servers. Participants build a set tailored to your project and leave it in the repository.
  3. Day 3. Review and final projectAI code review and its limits, then spec-driven development with GitHub Spec Kit. At PKO this was a loan calculator that went from an empty repository to production in one afternoon, with a new business requirement added just before the deadline.

The programme is built on Copilot because that is the tool most teams use. If you work with Codex, Claude Code or Cursor, the exercises carry over unchanged, and on request we can compare the tools on your code.

Results at PKO Bank Polski

20 / 20participants deployed a working application on day three
3.7 → 7.6average knowledge-test score out of 10, before and after
33% → 2%“I don't know” answers in the same test

“AI in software development”, a three-day workshop for 20 developers at PKO Bank Polski, 21-23 September 2026.

What you keep

Instruction files and agents tailored to your repository, materials from every day, and the knowledge-test results from the start and end of the training, which you can show when accounting for your training budget.

We already have Copilot. Why do we need training?

A licence alone does not change how people work. At PKO Bank Polski the average knowledge-test score rose from 3.7 to 7.6 out of 10 in three days, in a team that already had the tools. The difference comes from instruction files, custom agents, code review habits and knowing when not to trust the assistant.

Can the training use our code? Will you sign an NDA?

Yes and yes. Exercises on your own repository work best, because people learn on the problems they will meet on Monday. I sign the NDA before I see any code, and we can run everything on your machines and licences.

Does it work with tools other than Copilot?

Yes. Instruction files, custom agents, writing the spec before the code, and review work much the same in Codex, Claude Code and Cursor. We agree on the call which tool we practise on.

How do you measure the effect of the training?

Participants take the same knowledge test on the morning of the first day and at the end of the last. The second measure is the final project: at PKO, 20 out of 20 people deployed a working application.

How big can the group be?

I have tested the programme with a group of 20. For a larger team we agree on the call whether to split it into groups.

How much does the training cost?

From PLN 7,000 net per training day, plus VAT, ordered directly from me with no training agency in between. The full programme is 3 days. We agree location and travel on the first call, and you get a fixed price before we start.

How to start

Book a short call and tell me who will be in the room and what your team struggles with most right now. I will tailor the programme to you, and you get a fixed price before we start.

Ask about training →

Prefer to send a brief first? Email [email protected]. Or message me on LinkedIn.

Has your team already built an AI application or agent and now has to decide what comes next? The production-readiness review may fit better than a training.