Time since we first saw it
Day 3
Similar jobs usually come down between day 1 and day 15.
From 1,368 similar postings that came down. Coming down is not always a hire.
Munich, Germany
Actively hiring, first seen by us 3 days agoBuild multi-agent harnesses that route tender documents and protocol archives into the right pipeline
Build a decision graph capturing why decisions were made, by whom, and how they played out
Reconstruct the reasoning behind decisions buried in messy German protocols
Own product features end to end, including the TypeScript and React frontend
We build multi-agent harnesses (specialized extraction, reasoning, and evaluation stages)
not just what was decided, but why, by whom, against which alternatives, and how it played out
The hard part is reconstructing the "why" when it's buried in a messy German protocol.
Since you own things end to end, that includes the frontend.
being comfortable not only with the AI engineering side, but with TypeScript and React too
You prototype fast and measure everything. You also care about reliability
so you read papers, build prototypes, and compare approaches instead of waiting for a best practice to show up
You reason from user pain to solution to measurable outcome, and you can sit across from a non-technical customer
Project memory as a compounding moat. We started with meeting transcripts. Now we're building a decision graph that grows with every project: not just what was decided, but why, by whom, against which alternatives, and how it played out. That graph feeds the next project. The hard part is reconstructing the "why" when it's buried in a messy German protocol. We want the person who finds that problem interesting. Stack. TypeScript, Next.js, Vercel, Supabase (Postgres + pgvector), LangChain, Vercel AI SDK, LangFuse, shadcn/ui. Every engineer gets €500/month for AI tooling: Claude Code, Cursor background agents, and whatever frontier model you want to test. No legacy. Greenfield. How we work
You reason from user pain to solution to measurable outcome, and you can sit across from a non-technical customer and understand how they actually work. Since you own things end to end, that includes the frontend. That means being comfortable not only with the AI engineering side, but with TypeScript and React too. You prototype fast and measure everything. You also care about reliability, because on a live construction project, wrong has consequences.
Actively hiringFirst seen by us 3 days ago, still listed, and not a repost of an earlier posting.
These dates show when we found the posting. Check the employer’s website to confirm it is still accepting applications.
Day 3
Similar jobs usually come down between day 1 and day 15.
From 1,368 similar postings that came down. Coming down is not always a hire.
1 open job
Open jobs by the week we first saw them, last 12 weeks.
Hiring recordCheck the employer’s posting for the current role and application details.
Already applied? Track this application
The problem
The construction industry loses $1.6 trillion a year to inefficiency, and most of it is avoidable. The knowledge that would prevent it already exists. It's just trapped in PDFs, meeting protocols, and email threads nobody can search. Every new project relearns what the last one already knew.
We're building the opposite of that: software for the people who run large, complex construction projects. It reads the documents, tracks the decisions, and catches the mistakes early. The model isn't the moat. The project memory is, and it compounds: every project makes the system better, and a competitor starting today is already years of projects behind.
Our software runs today on a live autobahn construction program and an S-Bahn transit program: multi-year timelines, hundreds of thousands of pages of specs, protocols, and site communications, and real consequences when we get an answer wrong.
We're pre-seed, led by Realyze Ventures, whose LPs include Zech and other large European construction groups. Co-investors: D11Z, the family office behind Aleph Alpha, and the CDTM Venture Fund, backed by 300+ CDTM alumni including the founders of Personio and Alasco and DeepMind's Technical Director. 25+ live customers.
What you'd work on
Two problems, both of which need state-of-the-art answers.
Agent harness engineering for construction documents. One summary doesn't fit all. "Structural risk" means something different in an RFI, a cost review, and a schedule reconciliation. We build multi-agent harnesses (specialized extraction, reasoning, and evaluation stages) that route a 400-page tender document or a protocol archive into the right pipeline with the right context.
Part of that is context compression: what the system should remember, forget, and surface, and at which decision point, when a project runs five years and touches 50 stakeholders. There's no clean top-k answer, so this gets solved in production, not in a library. If you've read what Anthropic and Vercel have written about agent harnesses and thought "yes, that's the hard part of shipping production agents," this is the job.
Project memory as a compounding moat. We started with meeting transcripts. Now we're building a decision graph that grows with every project: not just what was decided, but why, by whom, against which alternatives, and how it played out. That graph feeds the next project. The hard part is reconstructing the "why" when it's buried in a messy German protocol. We want the person who finds that problem interesting.
Stack. TypeScript, Next.js, Vercel, Supabase (Postgres + pgvector), LangChain, Vercel AI SDK, LangFuse, shadcn/ui. Every engineer gets €500/month for AI tooling: Claude Code, Cursor background agents, and whatever frontier model you want to test. No legacy. Greenfield.
How we work
Who we're looking for
You reason from user pain to solution to measurable outcome, and you can sit across from a non-technical customer and understand how they actually work.
Since you own things end to end, that includes the frontend. That means being comfortable not only with the AI engineering side, but with TypeScript and React too.
You prototype fast and measure everything. You also care about reliability, because on a live construction project, wrong has consequences.
You're comfortable without a map. Most of these problems don't have answers yet, so you read papers, build prototypes, and compare approaches instead of waiting for a best practice to show up.
Level is open. If you're exceptional, we'll build the scope around you.
Nice to have, not required: serious open-source work, an earlier stint as an early employee at a startup, real depth in document understanding or agent systems, or something you shipped that replaced hours of human labor.
We work in English. German helps for customer calls but isn't required. We have native speakers for that.
The offer
You'll be one of our first engineers, which means you own critical parts of the product end to end and work directly with the founders. It's early enough that you help decide how the company works, and the path from here goes to either top individual contributor or VP Engineering, depending on where you're strongest.
Equity: 0.5% to 1.5%, four-year vest, 1.5-year cliff.
Hybrid: three days a week in our central Munich office. If that's a dealbreaker for you, let's talk.
How hiring works
Three stages, no endless loops.
Offer decided within 24 hours of the case study, out within 48.
New AI Engineer postings at the employers we follow are up 340% over the last 6 weeks compared with the 6 before.
Each bar is one week of new postings. The darkest bars are the last 6 weeks.
Counted from the hiring pages we’ve followed for at least five weeks, so employers we add don’t look like growth. Updated weekly.
We file this job under AI Engineer, by the work the posting describes.
AI engineers build products on top of AI models that already exist, most often large language models: they connect a model to a company's data and tools and ship the assistants, agents and search features people use.
Postings ask for strong Python or TypeScript, experience with model APIs, retrieval and agent frameworks, and ways to test whether the model's answers are good enough.
An ML engineer trains and serves models; an AI engineer usually starts from a model someone else trained and builds the product around it, though titles often mix the two.
Selected passages from the saved posting. Check the full description for conditions and exceptions.
No pay amount identified in the saved description.
Munich, Germany
Equity: 0.5% to 1.5%, four-year vest, 1.5-year cliff. Hybrid: three days a week in our central Munich office. If that's a dealbreaker for you, let's talk. How hiring works
No clear work-authorization passage found. Eligibility is unconfirmed.
We couldn’t identify enough role detail in this saved description to support an AI answer. Read the full posting