About the Role As a science-focused Member of Technical Staff, you will curate the multimodal biological datasets that power our models, analyze model behavior, and ensure our model outputs meet rigorous scientific standards. You'll co-develop benchmarks, filters, and validation pipelines with engineering peers so biological world models remain trustworthy and actionable.
What You'll Do - Source, normalize, and steward large-scale genomic, epigenomic, transcriptomic, proteomic, and imaging datasets with rigorous metadata and provenance.
- Build evaluation suites and benchmarks that stress-test generative biological models across modalities and tasks.
- Partner with AI engineers to analyze model outputs, run ablations, and surface insights that guide future architecture and training improvements.
- Integrate new datasets and annotations from external collaborators while maintaining compliance, privacy, and ethical standards.
- Communicate findings and best practices across Radical Numerics so teams can trust and act on model results.
What We're Looking For- PhD in genetics, computational biology, or a related field, OR demonstrated experience in biotech with a strong track record of impact over 3+ years.
- Proven experience curating, harmonizing, and analyzing large biological datasets (e.g., genomics, single-cell, spatial, or imaging).
- Fluency with Python, data tooling, and reproducible workflows (git, notebooks, containers).
- Ability to interrogate model outputs, debug unexpected behaviors, and translate findings into actionable recommendations.
- Clear communicator who can bridge scientific context with engineering teams and partner organizations.
- Curiosity and resilience when tackling open-ended scientific challenges.
Nice to Have - Familiarity with generative model evaluation, red-teaming, or safety analysis in scientific domains.
- Experience with statistical validation, quality control, or benchmarking for scientific or ML systems.
- Experience building benchmarking frameworks or open datasets that became community standards.
- Contributions to shared analytics tooling or reproducible research pipelines.
Why Radical Numerics - Help produce the multimodal biological world models that will power rapid detection, response, and countermeasures across global health.
- Collaborative culture that values rigor, creativity, and cross-disciplinary partnership across AI labs, biotechs, hospital systems, and national research institutes.
- Competitive compensation, comprehensive benefits, and support for continual learning.
