Szymon Pacholski — AI Engineer

I show how I work with AI agents on real software projects.

I'm Szymon, an AI Engineer. I write about AI SDLC from practice: domain context, specs, tests and review when working with AI agents. What works, what doesn't, and what I don't know yet.

What Will the Agent Check Its Result With?

AI/LLMAI AgentsSoftware Engineering

Published: 2026-11-21

Before handing a task to a coding agent, I'd check what it can verify its result with. What the AWS Emulator story shows about this, and what I'd give the agent as the first task when there is no check yet.

Bringing AI Into a Team: People, Workshops, Processes and What to Measure

AI/LLMSoftware EngineeringProductivity

Published: 2026-11-14

How I'd bring AI tools into a team so people want to use them: start with those who already use AI, run workshops, make the process clear for whoever supervises the agent and measure time and quality of the delivered work.

Where the Time Goes After the Agent Writes the Code

AI/LLMSoftware EngineeringProductivity

Published: 2026-11-07

With a coding agent a new endpoint takes a few minutes, and then the change can wait a week for a decision. Where that time goes and what could shorten it.