Medicinal Chemist — AI Drug-Discovery Challenge Designer

Obsidian; New York (NY); Remote; USD 90,000 - 150,000; Full time

Benefits offered by this job

Claude Max subscription reimbursed

Job summary

Obsidian is seeking a domain-expert to design advanced evaluation problems for frontier AI in drug research. You will craft realistic, source-grounded data worlds and ensure every decisive value has a real provenance.

The role emphasizes rigorous reasoning and defense of conclusions rather than pattern matching. You will also develop grading materials, run thorough validations, and oversee reworks grounded in scientific scrutiny.

Qualifications

• PhD or equivalent industry track record in relevant fields.
• Ability to pose defendable problems that resist shortcuts.
• Experience judging reasoning quality, not only final answers.

Responsibilities

• Design the challenge and identify the forcing chain that makes conclusions inevitable.
• Build a world with data room from real, source-grounded data.
• Write a grading document that scores memo substance, not style.
• Validate the task with verification and calibration before submission.
• Revise the task if reviewers request changes.

Skills

Enzymology

Structure-guided design

Fragment-based discovery

Medicinal chemistry

Rigorous reasoning

Education

PhD or equivalent

Tools

Claude Code

Claude Max

Job description

Obsidian is seeking a domain-expert to design advanced evaluation problems for frontier AI in drug research. You will craft realistic, source-grounded data worlds and ensure every decisive value has a real provenance.

The role emphasizes rigorous reasoning and defense of conclusions rather than pattern matching. You will also develop grading materials, run thorough validations, and oversee reworks grounded in scientific scrutiny.