Role
We're looking for a senior AI Engineer for a consulting assignment focused on building AI-driven capabilities in a business-critical, quality-sensitive setting for a company within healthcare/medtech.
Responsibilities
- Build AI functions that turn clinical content into practical, actionable suggestions for therapists in their daily work.
- Design end-to-end pipelines, including prompt and chain architectures, with structured outputs broken into clear, independently testable steps.
- Own model selection and strategy, including fine-tuning open-weight models.
- Help establish a data foundation with measurable quality, ensuring data is de-identified, version-controlled, and reproducible.
- Define per-function evaluation methodology — identifying which error types matter most and what quality bar must be met before output reaches a human reviewer.
- Set up regression testing and continuous quality monitoring in production.
- Build traceability mechanisms capturing what the AI suggests versus what the reviewer approves or edits, supporting transparency and long-term improvement.
Required Skills
- Proven experience building LLM-based systems that run in production, not just proof-of-concept work.
- Able to give concrete examples of how you've iterated on and improved live solutions over time.
- Strong background across data modeling, pipeline development, and model-related engineering.
- Clear understanding of why versioning datasets matters as much as versioning code.
- Experience handling sensitive personal data, with a real grasp of how nuanced de-identification can be.
- Comfortable working against defined quality bars rather than optimizing for a demo.
- Understands the stakes of reliability in AI systems, particularly where mistakes carry real consequences.
- Consulting mindset: able to step into an existing team, collaborate closely with domain experts, and leave behind solutions others can maintain.
Nice to Have
- Background in medical technology, healthcare, or other regulated fields with strong traceability requirements.
- Experience with speech-to-text systems, particularly for Swedish as a lower-resource language.
- Familiarity with the .NET ecosystem.