Senior ML Engineer - Foundation Models for Drug Discovery

F. Hoffmann-La Roche AG; New York (NY); On-site; USD 193,000 - 358,000; Full time

Job summary

Genentech, a Roche subsidiary, seeks a Principal Machine Learning Engineer to join our Foundation Models team in AI for Drug Discovery. You will own the MLOps/AgentOps stack, architect autonomous agents that leverage tools, and drive multi-step reasoning across drug-discovery workflows.

We expect leadership in building large-scale ML systems, production Python/PyTorch codebases, and a roadmap for AI4DD platforms. Onsite presence is expected; relocation is not offered.

Qualifications

• PhD with 5+ years, MS with 8+ years, or BS with 10+ years of building and owning large-scale ML systems.
• Strong Python and software engineering fundamentals.
• Experience with distributed training and ML infra on AWS or HPC.

Responsibilities

• Own the MLOps/AgentOps stack for Foundation Models and establish evaluation harnesses.
• Architect and deploy autonomous agents that use tools and retrieve evidence across data sources.
• Design scalable training and inference systems and maintain production Python/PyTorch codebases.
• Define the long-term engineering roadmap and serve as ML infra lead.
• Translate research problems into efficient, shippable systems.

Skills

Python

Git

CI/CD

PyTorch

JAX

AWS

HPC

Agent orchestration

LangGraph

MCP-based tooling

Education

BS in Computer Science or related

MS in Computer Science or related

PhD in Computer Science or related

Tools

Git

PyTorch

JAX

AWS

Job description

Genentech, a Roche subsidiary, seeks a Principal Machine Learning Engineer to join our Foundation Models team in AI for Drug Discovery. You will own the MLOps/AgentOps stack, architect autonomous agents that leverage tools, and drive multi-step reasoning across drug-discovery workflows.

We expect leadership in building large-scale ML systems, production Python/PyTorch codebases, and a roadmap for AI4DD platforms. Onsite presence is expected; relocation is not offered.