Software Engineer (gn) @ Stealth AI Venture, Berlin

What we're building Factories run on machines that generate huge amounts of data, and almost none of it gets used. We're changing that. We build AI software that turns machine and operational data into knowledge people can act on: for troubleshooting, for decisions, and for automation. We're early and in stealth. Our two founders both held C-level roles at leading German robotics ventures. They have spent years bringing advanced technology onto real shop floors, and now they're hiring the founding team. The role You'll build the software that makes sense of what happens on a manufacturing floor. That includes sensor measurements, actuator states, quality measurements, and operational events, streamed from connected machines and autonomous systems. Our platform processes this data to power analysis and automated workflows. We work AI-first. Our coding agents know our platform and our engineering practices, and you'll use them every day. You're still the engineer, though. You'll review every line, understand what it does, and test it properly. That frees you to focus on the application, the industrial domain, and the quality of the system. We care about what you can show, how fast you learn, and whether you take ownership. Seniority and years of experience matter less. What you'll do • Own features end to end: understand the problem, choose the approach, then build, test, ship, and follow up • Get to know the existing codebase in depth, including its architecture, dependencies, data flows, and failure modes • Write code with AI tools and review each change line by line for correctness, maintainability, security, and impact on the wider system • Build meaningful unit, integration, and end-to-end tests and wire them into CI • Test against messy real-world data, using recorded streams, representative datasets, and simulated devices. Cover missing or delayed measurements, duplicate or out-of-order events, invalid values, and dropped connections • Reproduce failures, fix them, and add regression tests so they don't return. Make sure tests reflect the intended behavior, and challenge the assumptions in the implementation • Build and improve backend services, APIs, and data pipelines for industrial analytics, knowledge retrieval, and automated workflows • Learn how our users and their factories work, so you can spot the right improvements and make good engineering calls • Help wherever it's needed: integration, deployment, documentation, and technical investigation. Flag progress, capacity, trade-offs, and blockers early You might be a fit if you • Write strong, practical Python and understand the advanced features and patterns used in production code • Have built and maintained substantial software: finding your way in an existing codebase, debugging behavior you've never seen before, and shipping reliable changes • Think like a tester: you turn requirements into test cases, set sensible test boundaries, build useful test data, and dig into failures • Have solid fundamentals in software design, APIs, data handling, testing, debugging, and version control • Can explain what code does, why it works, where it could break, and how to verify it, including code an AI helped write • Like AI coding tools, use them a lot, and stay critical of their output, because you own the result • Break down unfamiliar problems, investigate on your own, and ask sharp questions when you need to • Find the next useful step yourself and keep moving without being told what to do • Are comfortable with shifting priorities, varying workloads, and a broad scope, and communicate and prioritize clearly • Write and speak clear English A degree is optional. We care about evidence that you can engineer, not about a title or a specific path. Bonus points Depth in one of these areas makes you a particularly strong candidate. We don't expect all three. • ML and industrial data: strong ML fundamentals and hands-on work with time-series, sensor, machine, or robotics data. You've dealt with data quality, evaluated models, and turned results into useful product behavior • Agents, autonomous systems, and industrial automation: agent coordination, workflow orchestration, state management, or failure handling. IIoT experience, such as telemetry pipelines, device interfaces, MQTT, or OPC UA, is a big plus • Knowledge graphs and retrieval: knowledge representation, graph database design, schema and relationship modeling, graph queries, or retrieval that connects structured knowledge to what the application needs Also useful: cloud deployment, containers, and testing software against simulators or physical devices. Why now You'll be one of the first engineers. Your code goes to production, and your decisions shape the product and how we build it. You'll work directly with the founders from day one. We'll tell you more about the company, the founders, and our backers in the first conversation. Find more English Speaking Jobs in Germany on Arbeitnow