- Farringdon, London, United Kingdom
- Hybrid
- Full-time · Mid
- 3+ years
A role focused on bridging the gap between patent expertise and AI product development by defining high-quality outputs, testing AI models, and performing human-in-the-loop evaluations.
Responsibilities
- Be the voice alongside product and engineering teams building our end-user experiences of the various types of end users who use patent analytics for business decisions
- Define what good looks like across a broad range of AI-assisted uses, such as portfolio exploration, comparison, competitor benchmarking, technology and sector trends
- Capture that definition in governed, versioned question sets, graded example answers, scoring rubrics, and acceptance criteria
- Perform structured human-in-the-loop evaluations and drive the eyes-on evaluation layer for our AI assistant and analytics features
- Collaborate closely with other patent experts in the team to develop methods that scale eyes-on human evaluation
Requirements
- Degree in a scientific, engineering, or technical subject
- Nice to have: patent analytics platforms
- Nice to have: AI evaluation
- Nice to have: rubric-based scoring
- Nice to have: LLM-as-a-judge