# Frequently asked questions about Stefan > Short, grounded answers to the questions visitors most often ask about Stefan Preu, his work, hiring him and this bot. > status: draft-owner-review · updated: 2026-10-20 · source: https://bot.unbrained.dev/knowledge/faq Short answers to common questions about Stefan Preu. Each answer points to the page with more detail. ## Who is Stefan Preu? Stefan Preu, who goes by Steve, is an agentic systems architect based in Austria. He builds tools and systems that help humans and AI agents work with durable context. His main project is the open-source PM CLI. More: the "About Stefan Preu" page and https://preu.at/about ## Is Steve Preu the same person as Stefan Preu? Yes. Steve is short for Stefan. It is the name on his GitHub profile and on X (@StevePreu). On GitHub his handle is unbraind; on npm it is unbrained. ## How do you pronounce Preu? Preu is German, /pʁɔɪ̯/, like the German word "treu". In English, just say "proy", which rhymes with "boy" or "joy". ## What does an agentic systems architect do? In Stefan's description, he designs the systems around AI agents: what they know when a session starts, which tools they can use, where the boundaries are, and how anyone can tell the work is done. The models come from the labs; the system around them is his part. ## Why is Stefan so interested in context? Because getting the right information to an agent when it needs it shapes the result. Goals, decisions, history and relationships all matter, and a fresh session should be able to find its way back into the work without someone rebuilding the background by hand. More: the engineering approach page. ## What is PM CLI? PM CLI is Stefan's open-source, MIT-licensed project management tool for people and coding agents, published on npm as @unbrained/pm-cli. It keeps goals, decisions, tasks with acceptance criteria and evidence where a fresh session can find them. He started it in February 2026 and is its only human maintainer. More: the PM CLI page, https://pm-cli.unbrained.dev and https://preu.at/work/pm-cli ## Is PM CLI git-native? Older descriptions call it that. Stefan's current direction is that PM CLI should be fully usable without a Git repository and that the SDK should support other storage backends. That is a direction, not a claim that it has fully shipped; check the current documentation at https://pm-cli.unbrained.dev. ## How should I try PM CLI? Decide first whether your project is new or already established. Then ask your agent how PM CLI could fit your existing workflow and tools, and try that fit rather than copying someone else's setup. Stefan is interested in what helped, what did not, and where you or the agent hit friction. ## What else has Stefan built? Agent-friendly command-line tools for Home Assistant, Jellyfin, SearXNG and Monica, extensions and companions around PM CLI, and an experimental Ollama model manager called LlamaLift. Forks on his GitHub are not his original work. More: the projects page and https://preu.at/work ## Do I have to copy Stefan's workflow? No. Stefan wants people to build their own workflows from useful parts, adapt tools without waiting for a feature from him, and not be forced to change how they work when new capabilities arrive. That is a design principle he works toward, not a claim that every part is finished. ## Can I hire Stefan, or work with him as a consultant? Stefan is open to engineering roles, consulting projects and collaborations, and gives hiring and consulting equal weight. Scope, availability and terms need a direct conversation. Contact him through https://preu.at/contact or at stefan@preu.at. ## What does Stefan charge, and is he available? This bot does not know. There are no published prices, rates, packages or availability commitments, and it cannot make any. Ask Stefan directly through https://preu.at/contact. ## How would working with Stefan start? With the work and knowledge you already have: your goals, decisions, how information moves and where people get stuck. Early questions he asks: do you already use LLM providers, which ones, and what access may agents have to company data? More: the working together page. ## Will we depend on Stefan afterwards? He aims for the opposite. It matters to Stefan that you can keep using and changing the result independently. Continued involvement should be valuable because the system keeps improving, not because it cannot run without him. ## What would make a collaboration unworkable? Stefan needs a legitimate way to experiment with relevant data. Confidentiality matters, and options such as anonymized subsets or local models may help where permitted. If every workable route is ruled out, there is little room for the kind of work he wants to do. ## Is Stefan moving away from programming? No. He describes himself as still programming, now largely through LLMs. Code remains important as something agents can inspect, return to and debug, and as a reference for future work. ## Does Stefan prefer MCP or CLIs? He generally prefers CLIs when tools should be combined in scripts and workflows, or when data should be filtered before it reaches the model. He also sees a place for a small, focused MCP interface. He watches what agents actually use and where they struggle, then adapts. More: the tools and agents page. ## What does Stefan check when an agent says it is finished? It depends on the task. He might take a quick look to understand the result, then ask whether it was tested end to end as a real user would use it. He wants to know what supports the claim that the result is good, not only that the agent says it is done. ## Should an agent ask questions before a long task? Stefan asks agents to get familiar with the project and its recorded context first, check what is already answered, then bring the questions that genuinely need his input, ideally with possible answers. There is no universal rule about when to interrupt later; it depends on the task. ## Where is Stefan based, and which languages does he speak? He is based in Lower Austria, Austria. German is his native language and his CV lists English as business fluent. ## Where does Stefan work right now? This bot's public knowledge does not cover Stefan's current employment. His earlier roles, as listed in his CV, are on the background page. For anything current, ask Stefan directly. ## How is preu.at related to unbrained.dev? Both belong to Stefan. https://preu.at is about him and his work; https://unbrained.dev presents services and client projects. bot.unbrained.dev hosts this AI representative. ## Am I talking to Stefan? No. This is steve's bot, an AI representative of Stefan Preu. It answers from curated public knowledge about him and can be wrong. It cannot book meetings, send messages or make commitments for him. To reach Stefan himself, use https://preu.at/contact. More: the "About this bot" page.