Senior Multimodal ML Scientist – Autonomous Lab AI

Roche; South San Francisco (CA); On-site; USD 168,000 - 312,000; Full time

Benefits offered by this job

Annual bonus

Job summary

Genentech is seeking a Senior Multimodal Machine Learning Scientist to lead hands-on development of intelligent autonomous-lab systems. You will design, train, and deploy multimodal models that fuse video, camera feeds, and rich telemetry to enable real-time decision-making and autonomous recovery across lab workcells.

The role emphasizes building end-to-end ML pipelines and evaluation suites, collaborating with scientists, automation engineers, and software developers to scale AI for drug

Qualifications

• 3+ years of hands-on experience deploying production ML systems.
• Strong background in Computer Vision and Vision-Language Models.
• Experience building multimodal models combining camera feeds with sensor data.
• Proven ability to build and maintain evaluation pipelines.

Responsibilities

• Develop and fine-tune state-of-the-art vision models for real-time QC across lab workcells.
• Integrate camera feeds with time-series telemetry and logs for robust failure analysis.
• Build predictive ML models from spectroscopy and chromatography data.
• Create end-to-end data pipelines: collection, labeling, training, deployment, monitoring.

Skills

Computer Vision

Vision-Language Models

Multimodal Models

PyTorch

OpenCV

Education

MS/PhD in Computer Science

Tools

PyTorch

OpenCV

torchvision

Job description

Genentech is seeking a Senior Multimodal Machine Learning Scientist to lead hands-on development of intelligent autonomous-lab systems. You will design, train, and deploy multimodal models that fuse video, camera feeds, and rich telemetry to enable real-time decision-making and autonomous recovery across lab workcells.

The role emphasizes building end-to-end ML pipelines and evaluation suites, collaborating with scientists, automation engineers, and software developers to scale AI for drug