Senior Machine Learning Engineer, Agentic Science/Generative Models, AI for Biology & Translation (AIBT)

About the role

A healthier future drives us to innovate, continuously advance science, and ensure access to healthcare for generations. Roche’s Research and Early Development organisations (gRED and pRED) leverage AI, data, and computational sciences to accelerate drug discovery and development. The Computational Sciences Center of Excellence (CoE) unifies these efforts, harnessing data and AI to deliver transformative medicines.

The AI Biology & Translation (AIBT) department within Genentech’s Computational Sciences Center of Excellence (CS-CoE) is building next-generation AI systems for biology. We develop AI models that learn from biological data at scale, generating insights into disease mechanisms, therapeutic opportunities, and human biology. We seek a highly motivated Senior ML Engineer to join our Generative Modeling team, contributing to the design, development, and scaling of large-scale foundation models and agentic systems for therapeutic discovery.

This role spans the full stack: agent design, orchestration logic, infrastructure, AgentOps, and MLOps to ensure robustness, reproducibility, and efficiency at scale. You’ll collaborate in a multidisciplinary environment with ML scientists, ML engineers, and computational biologists.

Responsibilities

• Agentic systems
  • Build agents that use tools, retrieve evidence, and reason across multi-step scientific workflows
  • Design reliable interfaces between agents and biological, genomic, and clinical data sources
  • Create evaluation harnesses to validate agent output against scientific ground truth
  • Implement self-improving and autonomous loops for autoML and lab-in-the-loop systems
  • Develop agent memory and context management for long-horizon workflows
• Models and production systems
  • Build, fine-tune, deploy, and scale foundation models and LLMs in production
  • Own production Python/PyTorch (or JAX) codebases to turn research ideas into reliable, reusable software
  • Manage the MLOps/AgentOps lifecycle: experiment tracking, evaluation, monitoring, reproducibility, CI/CD, and infrastructure-as-code (Terraform, Helm, Kubernetes)
• Collaboration
  • Work with research scientists to scope and ship systems addressing open-ended scientific problems
  • Raise the engineering bar across gRED and Roche

Requirements

• BS/MS in CS, ML, engineering, or a related quantitative field
• 5+ years of experience building and shipping ML systems in industry
• Excellent Python skills; strong software and data engineering fundamentals (Git, automated testing, CI/CD, documentation)
• Track record of leading technical projects end-to-end
• Comfort with ambiguity and close collaboration with scientists
• Strong problem-solving and communication skills
• Interest or experience in applying ML to scientific discovery (e.g., biology, chemistry, drug discovery), including working with domain-specific data and models

Preferred Qualifications

• Inference-time scaling and optimization: test-time compute, sampling/search strategies, model routing, batching, caching, latency/cost/quality tradeoffs
• ML infrastructure on AWS (EC2, S3, EKS, SageMaker), including distributed training and inference on HPC
• Evaluation systems for agentic applications with scientifically defined correctness
• Agent orchestration frameworks in production (LangGraph, MCP-based tool integration)

Pay

The expected salary range for this position in San Francisco is $168,100–$312,300, with actual pay determined by experience, qualifications, geographic location, and other job-related factors. A discretionary annual bonus may be available based on individual and company performance.

Benefits

This position qualifies for the benefits detailed at the provided link.