Materials Discovery AI Scientist

Logical Intelligence; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Logical Intelligence in San Francisco seeks an AI Research Scientist to advance materials discovery by turning ideas into new molecules. You will build AI systems that propose and evaluate candidates and work with partner labs so synthesis and measurements teach the models.

Join a cross-disciplinary team of chemists, physicists, and ML engineers to develop energy-based models, diffusion-style approaches, and rigorous evaluations that guide experiments and model improvements with real lab

Qualifications

• PhD or Master's with hands-on ML for scientific problems.
• Experience with energy-based or diffusion/score-based models.
• Knowledge of chemistry or strong willingness to learn.
• Experience designing evaluations that correlate with downstream performance.
• Strong publication or open-source track record.

Responsibilities

• Build and improve the AI systems behind our discovery loop.
• Develop energy-based models that search chemical space efficiently.
• Own model evaluation with benchmarks, uncertainty, and lab-prospective tests.
• Compare AI predictions with lab measurements and translate insights into model changes.
• Design experiments with partner labs to maximize information from each synthesis/testing round.

Skills

ML background

Computational chemistry

Physics knowledge

PhD or MS + experience

Energy-based models

Diffusion models

Generative models

Chemistry familiarity

Evaluation design

Interdisciplinary teamwork

Education

PhD

Master's + industry experience

Job description

Logical Intelligence in San Francisco seeks an AI Research Scientist to advance materials discovery by turning ideas into new molecules. You will build AI systems that propose and evaluate candidates and work with partner labs so synthesis and measurements teach the models.

Join a cross-disciplinary team of chemists, physicists, and ML engineers to develop energy-based models, diffusion-style approaches, and rigorous evaluations that guide experiments and model improvements with real lab