AI-first engineerCzechia, remote
I’m Petr Klíma. I’m building Grindhouse, a fully remote environment for running and supervising coding agents on your own machines from any browser. All of its code is agent-written, and nearly all of it was produced through Grindhouse itself. Before that I shipped agent platforms and marketplaces as a senior developer and CTO.
Selected work
Two agent products in production and a marketplace that learned to see. All work
A fully remote environment for running and supervising coding agents. The agents run on your own machine or a rented server; you direct and review the work from any browser, desktop or mobile. Every line of its code was written by agents, almost all of it through Grindhouse itself.
An agent platform that organises a salesperson’s whole working day, from account suggestions and coaching to live guidance on calls. I worked on the core agent loop, context handling, integrations with Salesforce, HubSpot, Salesloft, Gong, Slack, Gmail and calendars, billing for units of agent work, briefs and post-call automation.
A consumer marketplace I helped build twice: first as CTO building the team and architecture, then returning to ship AI features such as item identification and listing generation from photos, and virtual try-on.
How I work
Implementation is no longer the slow part, so my attention goes to defining the work, reviewing it, and testing it independently. The full process and tooling
Explore the problem, the existing system and its constraints, and gather the information the implementing agent will need. Most failed agent runs trace back to this step being skipped.
Split higher-impact work into small end-to-end increments that can be exercised before the next one expands them.
Agents write the code. I read the changes for defects, unnecessary complexity, and structures that do not fit the system.
A separate agent uses the application through browser or computer-use tools. It can inspect code and logs when blocked, then reports what it found.
Clear issues go back to the development agent with context. Ambiguous reports get a second model’s opinion before I decide what changes.
Contact
I’m happy to compare notes, review an architecture, or talk about working together.
