AI for Science Engineer

Partner with research labs across ETH to identify high-impact use cases for agentic AI and co-develop novel pipelines and tools. Build planning and orchestration components that turn research goals into executable workflows under constraints such as cost, latency, and fidelity, and adapt as new results come in. Make what the system does inspectable: traceable runs, reproducible results, and interfaces that let researchers understand and steer the work. Support research teams with data preparation and curation, and help them optimize, specialize, and deploy models — fine-tuning, distillation, quantization, and efficient inference. Contribute to the evaluation and benchmarking of AI systems on scientific tasks, including open and closed frontier AI models. Develop and open-source reusable recipes, reference architectures, and templates that lower the barrier for others to build on. Stay at the frontier of agentic AI: evaluate emerging frameworks, prototype new patterns, advise research teams. Contribute to teaching and student projects.

The ETH AI Center is one of the largest AI centers in Europe with over 130 Professors from all 16 departments of ETH Zurich and over 1500 associated PhD students & post-docs. Our mission is to foster interdisciplinary AI research and pave the way from science to society.

• Preferably a doctorate in computer science or a computational discipline; a master's degree with strong relevant experience is equally welcome. Our main counterparts are researchers, so being able to engage with them on their own terms matters.

• Experience building and deploying agentic or LLM-based systems end to end, not only using them - typically 1–3 years of relevant experience, though exceptional recent graduates are encouraged to apply too.

• Strong software and systems engineering skills: services, APIs, data pipelines, containers, CI/CD.

• Hands-on experience with modern AI/ML tooling (LLM APIs, agent frameworks, RAG pipelines, PyTorch, HuggingFace) and with model optimization and efficient inference.

• Experience with GPU computing and high-performance or distributed computing environments.

• Exposure to planning, scheduling, workflow engines, or reasoning systems, and to scientific computing workflows, is an advantage; open-source contributions are a strong plus.

• An effective communicator who bridges researchers and engineers — proactive, delivery-oriented, and keen to enable others to do their best work.

• Fluency in English.