Software Engineer (gn) @ Industrial AI Venture, Berlin

This is an Atlantic portfolio company. About Marentis Marentis develops AI-supported software for manufacturing, robotics and complex automated machinery. We turn machine data and operational information into useful knowledge for troubleshooting, decision-making and automation. Our work connects sensor measurements, machine and robot states, quality data, alarms and expert knowledge. These problems require understanding both the data and its context: a fault may originate elsewhere in a production line, and similar symptoms can have different causes. We are a small, early-stage startup where individual contributions directly shape the product. We value curiosity, initiative and people who enjoy understanding difficult problems and making things work. The Role We are looking for an experienced, hands-on Software Engineer with strong Python skills and sound technical judgment. You will own backend features from problem definition through implementation, testing, deployment and maintenance. After onboarding, you should be able to investigate unfamiliar production code and independently review Python/backend pull requests: understand intended behavior, question assumptions, assess tests and failure paths, and identify effects on the wider system. We actively encourage AI-assisted development. You remain responsible for understanding, verifying and maintaining the resulting software. We value demonstrated delivery, ownership and learning ability more than a fixed number of years in the industry. What You Will Do • Build backend services, APIs and SQL-backed data pipelines for industrial analytics, knowledge retrieval and agent workflows. • Read and improve existing code, review pull requests, and make informed decisions about interfaces, data models, state and failure handling. • Design meaningful unit, integration and end-to-end tests, supported by continuous integration. • Test with representative machine data, recorded streams and simulated device inputs, including missing measurements, duplicate or out-of-order events, invalid values and connection failures. • Reproduce bugs, inspect whether test assertions verify the intended behavior, and add useful regression coverage. • Understand manufacturing users and equipment, contribute across integration and deployment, and communicate trade-offs and blockers early. What You Bring • Strong practical Python and backend engineering skills, supported by software you personally shipped and maintained. • Independent code-review ability: you can investigate a change, assess correctness, maintainability and security, and justify whether it is ready to merge. • Practical system-design judgment across APIs, SQL, database design, data validation, state management and concurrency where relevant. • Strong testing and debugging skills: you choose useful test boundaries, construct representative data and investigate failures. • Experience making imperfect operational, machine, sensor or similarly complex data trustworthy, with an interest in its industrial meaning and consequences. • Resourcefulness: you learn unfamiliar code and technologies through documentation, small experiments, focused questions and feedback. • Clear written and spoken English, constructive collaboration and reliable follow-through. We welcome different educational and professional paths. Direct manufacturing experience is especially useful; equivalent engineering depth from adjacent domains is also relevant. Particularly Relevant Experience Bring depth in one or more of these areas; expertise in all of them is not expected. • Machine data, analytics and machine learning: Time-series or sensor-data processing, curation, feature engineering, anomaly detection, condition monitoring or quality prediction. Explain how you handled timestamps, units, asset identity and data quality, and evaluated whether the result was useful. • Knowledge graphs and retrieval: Entity and relationship modeling, graph or relational database design, provenance, versioning and conflicting information. Experience combining identifiers, aliases, lexical search and semantic retrieval, with evaluations of relevance and applicability. • Agent systems and workflows: LLM tools, multi-agent coordination, orchestration, context and memory management, validation and failure recovery. Show how you evaluated quality, cost and latency, enforced permissions and chose appropriate execution boundaries. • Industrial integration, robotics and automation: Connecting software with production equipment, robot cells or automated machinery through telemetry, device interfaces or protocols such as MQTT and OPC UA. Experience with simulators, physical-device testing and equipment-specific behavior is valuable. • Cloud deployment, containerization, observability and operating software across customer environments are also useful. Find Jobs in Germany on Arbeitnow