Bio-Model Interpretability Engineer

Output Biosciences; New York (NY); Presencial; USD 150.000 - 210.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Equity in startup

Competitive salary

Medical, dental, and vision coverage

Startup benefits

Descripción de la vacante

Output Biosciences is advancing a biological reasoning model to reveal its learning about biology and molecular interactions. You will develop methods to probe representations, design experiments, and build interpretable tools for researchers and downstream systems.

You will work with pretraining and generation teams to feed insights back into model development, owning pipelines from probing experiments to production-ready interpretability tools on distributed infrastructure.

Formación

• PhD or Master’s/Bachelor’s with substantial research/engineering in model interpretability or representation analysis.
• Strong publication track record in top-tier venues (NeurIPS/ICML/ICLR) is preferred.
• Hands-on experience analyzing internal representations of large neural networks.
• Proficient in Python and PyTorch; comfortable with GPU workloads.
• Experience turning interpretability findings into reusable tools and production-quality code.

Responsabilidades

• Develop methods to probe and reverse-engineer the model’s learned representations across molecular scales.
• Design experiments to characterize capabilities and map learned biology to practical outputs.
• Build tools to extract biological understanding as explicit outputs for downstream use.
• Create interfaces connecting model internals to biology concepts for scientists.

Conocimientos

PhD in CS/ML/Physics/Math

Post-doc or industry research

Publication record in top venues

Experience analyzing model internals

Python

PyTorch

Production-grade code

Version control & code review

Experiment design for interpretability

Educación

PhD in CS/ML/Physics/Math

Bachelor’s or Master’s with 5+ years relevant experience

Herramientas

PyTorch

Python

GPU infrastructure

Git / version control

Descripción del empleo

Output Biosciences is advancing a biological reasoning model to reveal its learning about biology and molecular interactions. You will develop methods to probe representations, design experiments, and build interpretable tools for researchers and downstream systems.

You will work with pretraining and generation teams to feed insights back into model development, owning pipelines from probing experiments to production-ready interpretability tools on distributed infrastructure.