Head of Engineering for funded startups in Berlin

What a Head of Engineering does after a funding round

A Head of Engineering leads the people who build the product: hiring, onboarding, the release process and the quality of the code. After a round the money turns into hires, and each new developer needs access, context and a first task before writing useful code.

Berlin startups hire across borders. In the German Startup Monitor 2025, 42 percent of startup employees in Berlin came from abroad, against 32 percent across Germany (Munich Startup, 2025-10-03). A team like that works in English and needs its rules in writing. Where the team runs coding agents such as Claude or Codex, the agents need the same written rules.

E-commerce and real estate: what I built

At Valentins, then owned by Burda Digital, I was Project Lead and Developer. I led projects and programmed front ends and back ends for at least five e-commerce sites, among them shops for Lidl and REWE with millions of transactions (counts from 2011 to 2016), and I ran their conversion optimization. The news channel n-tv named Valentins the best online shop for ordering convenience. Details: Experienced in conversion optimization.

For a global real estate database I was developer, managing director and CTO. In its first years I managed the product and programmed the websites and databases myself. Details: A global real estate database. For another real estate database I built the near-real-time search on Elasticsearch. I was also editor in chief of an information portal on construction projects, which reached more than one million readers a year by 2024. Details: An information portal.

Onboarding checklist for developers and coding agents

A new developer and a new coding agent need much of the same on the first day. The list puts both side by side. Bring it to our call, and we mark what your team already has.

Item	For a developer	For a coding agent
Access	Accounts for the repository, the tickets and the chat on day one	A token scoped to the development environment
Data	Access to production data only where the task needs it	A database role that reads and cannot write
Rules	A written guide to branches, reviews and releases	The same guide, in the rule file the agent reads
Context	Architecture notes and the decisions behind them	A knowledge graph of code and documentation, queried before a change
First task	A small change that reaches production with a review	A small change that passes the CI checks before the push
Review	A named reviewer for the first merge requests	A hook that blocks forbidden tool calls before they run
Production	Releases only through the pipeline	No push to the release branch; a person confirms each production change

The agent column comes from a client’s platform on which several agent sessions and the human team work in one repository. Details: AI agents under the same rules as humans.

Who I am

I am Michael Wutzke, and I live in Frankfurt. Today I help a client break up a monolithic platform with Claude and Codex and teach practical work with Claude at Claude Hacker House. I am also interested in open-source AI models that a company runs on its own servers. Earlier I led teams as managing director of a global real estate database and as CIO of a Frankfurt fintech. Details: Career stages.