Benchling
Research Engineer, Model Evaluation and Improvement
San Francisco, CA. Posted 9 days ago
The role
You will build datasets, evaluations, and systems to improve frontier AI models for science. You will work at the intersection of software engineering, biology, and frontier AI to turn scientific work into rigorous tasks for models.
What you will do
- Build datasets for evaluating and improving frontier models
- Analyze model failure modes by running experiments across frontier models
- Build scalable data infrastructure and pipelines to curate and validate scientific data
- Collaborate with frontier AI labs to develop and evaluate new approaches
- Work with scientists to translate expert judgment into evaluation criteria
Required qualifications
- 2+ years at the intersection of biology and AI
- Academic or industry experience evaluating and improving scientific models or LLMs for biological applications
- Experience building with LLMs
- Curiosity and excitement about frontier AI
- Comfort working on ambiguous problems
- Collaborative mindset
- Desire to work in a fast-paced environment
About the company
Benchling is the AI platform for biotech R&D. Over 200,000 scientists around the world use it to design experiments, capture structured data, and run AI agents and models.