Senior ML Engineer – Agentic AI

About Kaiko kaiko.ai is building a next-generation agentic clinical AI assistant that helps clinicians reason across patient data, guidelines, and diagnostics.

Healthcare decisions are rarely made by a single person or from a single data source. kaiko’s assistant maintains longitudinal patient context across encounters, clinicians, and institutions, enabling collaboration, second opinions, and complex diagnostic workflows. The system is designed to operate safely in real clinical environments, with human oversight, auditability, and regulatory alignment at its core.

Our assistant core supports broadly applicable clinical tasks such as patient data navigation, guideline interaction, multimodal interaction (chat and voice), and care coordination. On top of this foundation, we are developing specialized diagnostic agents in areas such as oncology, radiology, and pathology.

We build in close collaboration with leading hospitals and research centers, including the Netherlands Cancer Institute (NKI). kaiko is a well-funded company with a growing international team, operating from Zurich and Amsterdam.

About The Role You will join Kaiko’s ML Engineering team building the agentic system, the software harness that turns powerful models into something clinicians can rely on in real workflows.

In healthcare, this matters more than anywhere else. Doctors don’t need another interface that produces fluent text. They need a system that supports structured clinical thinking and collaboration, synthesizes messy context, makes uncertainty explicit, and produces artifacts that can be inspected, discussed, and improved with expert feedback.

As a Senior ML Engineer, you will design, ship, and evaluate the harness components that make our agentic system safe and effective. You’ll contribute to reliability, evaluation, and the engineering that brings agents into clinical practice.

You’ll be based in Zurich or Amsterdam, with an expectation to spend around half your time in the office.

You will help build and evaluate:

• context lifecycle management from intent to verified, persistent outputs
• tools and integrations across internal systems and external data sources
• durable memory and state that support long-running, multi-step clinical work
• evaluation and verification loops that reduce drift, context loss, and hallucinations

About You

• Strong Python skills and solid Git collaboration experience (PRs, branching, code review)
• Experience building LLM-driven features such as prompt and context design, RAG-based retrieval and grounding, tool use, and practical failure handling
• Experience building agents with a clear view of what makes them succeed or fail including state management, planning and execution tradeoffs, tool reliability, and guardrails
• Experience with at least one agentic framework such as LangChain, AutoGen, or similar. Or experience building custom production-ready agentic systems
• Strong ML foundation with particular strength in transformers and how LLMs and VLMs behave in practice including capabilities, limitations, and evaluation
• Experience designing evaluation for LLM and agentic systems including deterministic test suites, scenario-based evaluations, and rubric-based or LLM-as-judge approaches
• Clear communicator comfortable with design discussions, code review, and cross-functional collaboration

Nice to have:

• Experience with knowledge graphs or structured representations for reasoning and retrieval
• Experience using workflow orchestration tools such as Dagster or similar
• Familiarity with distributed execution frameworks (e.g., Ray) and scaling workloads cleanly
• You stay up to date with the latest developments and literature on agentic systems and can turn new ideas into shippable engineering
• Some experience with Reinforcement learning

We are excited to gather a broad range of perspectiv