Fullstack AI Engineer

Preconstruction is the foundation of every project, where the work gets chosen, priced, and put together. But while the projects have grown, the documents multiplied, and the requirements gotten heavier, the way the work gets done looks almost the same as it did decades ago.

Arctis AI is transforming how preconstruction works, making the people who run it faster, sharper, and able to take on more than ever before.

We're a small, talent-dense team out of TUM, with backgrounds spanning Bain, KPMG, SAP, Snowflake, and AWS. We're backed by PT1, EWOR, and operators and investors from European construction and tech, including the owner of PERI and partners at Atlantic Labs and La Famiglia.

About the Role

As Fullstack AI/LLM Engineer, you'll build the agentic systems behind bid comparison, clause extraction, and risk detection, on top of large volumes of unstructured construction data.

What You'll Work On

• Deployments: Run each rollout end-to-end - data onboarding, agent configuration, pipeline setup, launch. When something blocks go-live, it's yours to clear.
• Data & integrations: Get customer data into shape - past projects, prices, documents, sub lists and connect Arctis to the systems precon teams use.
• Prototyping: Build fast prototypes that solve emerging customer problems, then work with the core team to turn the good ones into product.
• Feedback loop: You'll see how the product holds up in the field before anyone else. Bring what you learn back into the roadmap - which failures repeat, which requests keep coming up, what to build next.

Who We're Looking For

B.Sc. / M.Sc. in computer science, AI, or a related field

Experience shipping AI-powered products with LLMs in production, not just prototypes or demos

Strong backend fundamentals and comfort owning a modern web stack end to end

Hands-on experience with RAG pipelines and agent orchestration, including the hard parts

Experience defining and running real evals for AI systems

Strong product sense and enjoyment of working directly with customers

High agency: fixing problems, not flagging them and waiting

Readiness to work onsite in Munich

What You Get

Founder-level ownership of the entire deployment motion.

Direct customer contact. You'll join calls and feedback sessions and shape the product from what you hear.

Steepest possible learning curve. From week one you're on live customers, working directly with the founders. You won't be handed tasks, you'll be handed problems, and how you solve them is up to you.

Skin in the game. €75-100k salary + 0.5-1.5% equity

Whatever tools you need. Hardware, Claude Code, unlimited AI credits, productivity setup, just ask.

EGYM Wellpass, team dinners, and offsites.