Biotech AI Research Engineer: Evaluation & Improvement

Benchling, Inc.; San Francisco (CA); On-site; USD 136,000 - 167,000; Full time

Job summary

Benchling, Inc. is building an AI-centric platform for biotech R&D and seeks a role at the intersection of biology and artificial intelligence.

You will craft datasets, design evaluation tasks, and partner with scientists to turn complex biology into rigorous benchmarks for advanced models. You will work with frontier AI labs, translate expert judgment into measurable criteria, and help push models toward reliable scientific reasoning.

Qualifications

• 2+ years at the intersection of biology and AI, evaluating and improving scientific models or LLMs for biological applications.
• Experience building with LLMs, intuition for strengths and weaknesses, and designing systems around capabilities.
• Curiosity and excitement about frontier AI and understanding how to push model capabilities.
• Comfort working on ambiguous problems in a fast-changing landscape.
• Collaborative mindset to work with engineers, scientists, and external partners.

Responsibilities

• Build datasets for evaluating frontier models, turning complex scientific data into high-quality tasks for LLMs.
• Analyze model failure modes across frontier models to identify improvements.
• Create scalable data infrastructure and pipelines to transform scientific data into evaluation tasks.
• Collaborate with frontier AI labs to develop and evaluate new model-improvement approaches.
• Translate expert scientist judgment into problems and evaluation criteria that distinguish strong model behavior.

Skills

Biology & AI intersection

LLM evaluation

Model failure analysis

Collaborative work

Ambiguity handling

Experimentation mindset

Job description

Benchling, Inc. is building an AI-centric platform for biotech R&D and seeks a role at the intersection of biology and artificial intelligence.

You will craft datasets, design evaluation tasks, and partner with scientists to turn complex biology into rigorous benchmarks for advanced models. You will work with frontier AI labs, translate expert judgment into measurable criteria, and help push models toward reliable scientific reasoning.