Generative ML Scientist – Molecular Design & Biotech

Output Biosciences; New York (NY); On-site; USD 150,000 - 210,000; Full time

Benefits offered by this job

Competitive salary and equity

Excellent medical, dental, and vision

Ownership of day-to-day management

Creativity and growth-focused culture

Job summary

Output Biosciences is seeking a leader to design and implement generative models that operate across molecular modalities such as small molecules and peptides. You will architect, train, and evaluate systems that produce novel, biologically grounded molecules.

You will own the end-to-end training pipeline on distributed GPU clusters, focusing on models that integrate biological reasoning and meet real-world chemical constraints. Strong publication and coding practices are expected.

Qualifications

• PhD in CS/ML/related field with 2+ years post-doc or industry research, or 5+ years hands-on research experience.
• Strong publication record in top venues (NeurIPS/ICML/ICLR) is required.
• Proficient in Python and PyTorch; experience with distributed multi-GPU training.

Responsibilities

• Design generative architectures for molecular data across modalities (small molecules, peptides, mini proteins).
• Develop training approaches to learn from diverse biological signals and enable novel structures.
• Build controllable generation methods respecting chemical constraints and biology.
• Own end-to-end training: experiment design, distributed training, hyperparameter tuning, iteration.

Skills

Python

PyTorch

Distributed GPUs

Research leadership

Rigorous experimentation

Education

PhD in CS/ML/physics/math or related

Bachelor/Master with 5+ years research experience

Tools

Git

Open-source contribs

Job description

Output Biosciences is seeking a leader to design and implement generative models that operate across molecular modalities such as small molecules and peptides. You will architect, train, and evaluate systems that produce novel, biologically grounded molecules.

You will own the end-to-end training pipeline on distributed GPU clusters, focusing on models that integrate biological reasoning and meet real-world chemical constraints. Strong publication and coding practices are expected.