ML Engineer, Agents & Reasoning

About the Role

We are a seed-stage deeptech startup at the intersection of AI, robotics, and materials science, building an advanced platform that dramatically accelerates the discovery of new materials — particularly for the energy sector. Our work combines physics-informed AI, autonomous laboratory systems, and rich multi-modal experimental data to compress decades-long R&D timelines into years.

As an ML Engineer, Agents & Reasoning, you will design and build the agentic AI systems that sit at the heart of our materials discovery workflows. You'll turn predictive models into reliable, operational decision-making agents that work alongside physical experiments, robotic systems, and scientific datasets. This is a high-ownership, end-to-end role on a small, cross-functional team of ~12–60 people based in Berlin, Germany (on-site).

Please note: visa sponsorship is not available for this role.

Responsibilities

• Design and implement agentic systems that plan, reason, and act across real materials discovery workflows
• Build decision-making systems that operate over experiments, simulations, and scientific datasets
• Select next actions under uncertainty and encode when autonomy should act versus when a human should stay in the loop
• Implement planning, control logic, and uncertainty-aware decision-making tailored to physical systems and lab environments
• Encode operational, experimental, and safety constraints directly into agent behavior
• Define stopping criteria, fallback strategies, and recovery mechanisms to prevent brittle behavior
• Collaborate with AI researchers to embed predictive models into agent workflows and translate model outputs into executable actions
• Integrate agents with laboratory automation and software systems so agent outputs drive real-world actions
• Instrument agents with logging, monitoring, and diagnostics to support observability and debugging
• Build evaluation frameworks that assess decision quality, learning efficiency, and system behavior — beyond simple model accuracy
• Analyze failure cases and iterate on system design based on real-world experimental outcomes
• Own systems end-to-end: from prototype through deployment and ongoing operation

Requirements

Required

• 4–8 years of experience building ML-driven or algorithmic decision-making systems in production or applied research settings
• Strong background in scientific or structured data modeling (rather than language-first or NLP-heavy systems)
• Experience with planning, control, optimization, probabilistic reasoning, or decision-making under uncertainty
• Proficiency in modern ML frameworks such as PyTorch or JAX, paired with strong general software engineering skills
• Comfortable owning systems end-to-end, from early prototype through to reliable production operation
• Ability to reason clearly about system behavior in complex, partially observable environments
• Clear communicator who can collaborate effectively across AI, engineering, and scientific teams
• English fluency (additional language skills a plus)

Nice to Have

• Technical curiosity about physical systems, laboratory experiments, and real-world constraints
• Experience in materials science, chemistry, cleantech, or adjacent scientific domains
• Familiarity with laboratory automation or robotics integration
• Additional European language skills (German in particular)

Location & Work Arrangement

This role is on-site in Berlin, Germany. We work closely as a team in person, and we expect this role to be based full-time at our Berlin office. Visa sponsorship is not available.

Why This Role

• Work on genuinely hard AI problems at the frontier of scientific discovery and physical-world autonomy
• Join an early-stage, mission-driven team where your work directly shapes both the product and the culture
• Collaborate across AI research, engineering, and laboratory science in ways that are rare in a single role
• Contribute to technology with meaningful real-world impact in the energy transition and advanced manufacturing