Full Stack Data Engineer

As our Full Stack Data Engineer, you’ll own the data that powers everything we do - the flow of cement and concrete plant data from raw, messy ingestion through cleaning, to the pipelines our machine-learning recommendations run on. The split is roughly 50% data engineering, 50% backend, and it’s refreshingly hands-on with customers: you’ll dig into real-world data, explain what you find to non-technical people, and make getting data into alcemy dramatically easier.

What you’ll do

• Build and own the pipelines that ingest and clean messy plant data into something our ML can use
• Help us move from strict, structured data requirements toward flexible, AI-powered ingestion - so customers onboard with far less friction
• Track down data anomalies and explain your findings to customers in plain language
• Build and maintain the APIs and PostgreSQL databases behind our products, and keep production running smoothly
• Work shoulder to shoulder with customers, sales, and customer success - translating between the technical and the non-technical, * No lone warriors - you’ll work collaboratively alongside senior and junior engineers, owning features end to end: build, test, review, deploy.
• Small team, direct customer exposure, fast startup pace - so async communication, sharp prioritisation, transparency, and direct feedback are the norm, and how we work is something we shape together, and keep improving.
• We use Claude Code responsibly - AI to move fast, engineers always in charge and accountable, no AI-slop merge requests.
• Hybrid by design (Berlin Mitte office, up to 2 days remote, plus offsites and boat trips), with real room to grow along technical-expert or management tracks - and the chance to lead a product area within a couple of years., * Backend & data: Python throughout - FastAPI (auto-generated Swagger/OpenAPI), SQLAlchemy, pydantic, pytest, mypy; pandas, polars, numpy, sklearn; uv for dependencies
• Infra: PostgreSQL, AWS on Kubernetes managed with Terraform, ArgoWorkflows for scheduled jobs, GitLab for code
• Frontend: React & Next.js in TypeScript with TailwindCSS; Blueprint.js / Radix UI; TanStack Query or axios; Streamlit for internal tools

• You love getting to grips with an unfamiliar codebase quickly, and you’re eager to start delivering work that matters from the start
• You’re comfortable working directly with customers, sales, and customer success - patient, plain-spoken, and at ease with non-technical enterprise teams
• You’re open to using AI tools to augment how you work

Nice-to-haves

• Experience with Flask or Django and Fast API
• PostgreSQL or similar SQL databases
• Some exposure to machine learning, Note: We don’t currently offer visa sponsorship, so you’ll need a valid EU work visa.

• 25 days’ vacation - plus an extra day for every 2 years here, and the days between Christmas and New Year off
• €1,500/year and 3 days/quarter for your own development
• The Mac of your choice
• An office in the heart of Berlin Mitte

alcemy is on a mission to keep 100 million tons of CO out of the atmosphere every year by 2030. Making concrete accounts for around 8% of global CO emissions, and our machine-learning software runs inside cement and concrete plants 24/7 to cut it - without sacrificing quality or driving up costs. We’re already live in a third of Germany’s cement plants and 30+ ready-mix concrete plants.