~/jiri-studnicka $ whoami

Agentic Engineer

I'm Jiří Studnička. Twenty years building backends for banks, insurers and energy companies. Today I build systems where AI agents write, test and ship production software — with a clear brief, verified results and human accountability.

// production software since 2004 · agentic systems since 2023

Empty bucket. Lower it to the water.Drag ↕. Drag elsewhere to rotate the well.Use arrow keys for bucket height, Home to lower it into the water, or End to raise it. Click the full bucket at the top to pour.

A cross-section of a well: a windlass above layers of stone, with water below.
Studnička. Beneath the surface.

// depth −7 m · sediments of practice

Experience runs deep

Studnička is Czech for a little well. It took twenty years of enterprise strata to dig — and that is exactly the depth every agent I set loose on production code draws from.

  1. 2004

    First lines in production

    IIS Tábor — Gupta and .NET. The craft begins.

  2. 2008–15

    The .NET era & a first company

    Capgemini, Minerva, RADIUM. Co-owner and team leader at 4Leaders.

  3. 2015–18

    Leveris — banking startup

    Scala, Akka, Play, Kafka, MongoDB. Functional programming, monads, event sourcing. A bank built from scratch.

  4. 2018–19

    E.ON SolarCloud

    Scala and Akka Persistence for the energy sector — event sourcing in live production.

  5. 2019–20

    Home Credit — big data

    Spark, Kafka, HBase, Hadoop, Hive. Integrating the primary systems and DWH of a multinational group.

  6. 2020–22

    Generali, SOFTINO

    Java, .NET, Python. Spring Boot, Oracle, Azure Synapse, Spark.

  7. ~2023

    GPT-3.5 — a turning point

    I understood the craft had just changed. I've been doing agentic engineering full-time ever since. Experience carries forward — you reforge it.

“My job is to give agents a clear brief, preserve context and verify that the result holds up in production.”

Claude CodeCodexClineRoo CodeAugment CodeOpenCode

// depth −18 m · the tool seam

The Smithy

I named my tools after the smithy. Not for the romance — because agentic engineering is a craft: it has a process, tools and quality control. This is my production cycle.

01 · the anvil

Anvil

issue → spec

Checks the brief against real code. Finds ambiguities and missing context before implementation begins. A human approves changes to the brief.

  • read-only analysis
  • human-in-the-loop

02 · the forge

Forge

issue → pull request

Coordinates agents through implementation, code review, tests and CI. Tracks requirements, manages fixes and verifies the result after merge.

  • code review
  • tests and CI
  • requirement ledger

03 · the quench

Temper

PR → proof

Tests in a real browser and captures each step. When the interface changes, an agent helps repair the scenario for another repeatable run.

  • screenshot of every step
  • self-healing scenarios

CogneraSpec

Links requirements to code and tests. Each has its own REQ-ID and verifiable evidence. Gaps between the brief and the result become visible.

AI Toolkit

Commands and skills for Claude Code and Codex. Repeatable workflows for reviewing and improving security, performance and accessibility.

Meeting Agent

Turns meeting recordings into GitHub issues in the context of the repository. Includes what was said and what appeared on screen.

…and more: agent-system (orchestration via Telegram), MCP servers, invoiceAI.

01

Spec-first. An agent without a proper assignment is a random code generator.

02

Proof, not vibes. I connect every requirement to tests and verifiable evidence.

03

Human-in-the-loop. AI writes. An engineer stays accountable.

// depth −29 m · load-bearing masonry

Where it all stands

// depth −42 m · the waterline

At the bottom of every well there is water

Have a project you want to bring AI agents into? Tell me what you are building, where development gets stuck and what needs to change.