2 days ago
Base Salary
$168k - $312k/yr
Responsibilities
- Develop and fine-tune vision models for real-time visual quality control in laboratory workcells.
- Build multimodal models combining camera feeds, video, sensor telemetry, system logs, assay metadata, and other laboratory data.
- Develop predictive models using spectroscopy, plate-reader, and chromatography data to support assay quality and closed-loop experimental decisions.
- Create end-to-end data and ML pipelines covering data collection, curation, labeling, training, deployment, monitoring, and improvement.
- Build agentic pipelines that interpret laboratory data, reason through workflows, and execute autonomous recovery decisions.
- Design evaluation suites to benchmark model and agent accuracy, robustness, and decision safety.
- Deploy and optimize scalable, low-latency ML inference and monitoring infrastructure for diverse models, instruments, and scientific applications.
- Collaborate with scientists, automation and robotics engineers, and software engineers to translate scientific requirements into practical ML systems.
Requirements
- MS or PhD in Computer Science, Data Science, AI, or a related quantitative field.
- At least 3 years of hands-on experience deploying production ML systems or translational research models.
- Strong computer vision and Vision-Language Model experience, including training, fine-tuning, or evaluating visual perception architectures.
- Strong foundations in machine learning, statistics, and data analysis, with experience developing and deploying predictive models.
- Experience building and deploying multimodal models that combine image or video streams with sensor logs, time-series data, or tabular scientific data.
- Hands-on experience building, orchestrating, or tuning agent harnesses.
- Demonstrated mastery of evaluation pipelines for benchmarking model and agent accuracy, robustness, and safety in production environments.
- Ability to collaborate across scientific, robotics, automation, and software engineering disciplines.
- Preferred experience with synthetic vision data generation, video anomaly detection, edge-vision deployment, lab automation hardware, analytical instruments, or industrial IoT telemetry.
- Preferred exposure to Bayesian optimization, active learning, reinforcement learning, or Design of Experiments for closed-loop experimentation.
- A public GitHub or GitLab portfolio featuring production ML pipelines, computer vision projects, agentic frameworks or evaluations, or multimodal models is preferred.
Benefits
- Benefits are provided through the company benefits program linked in the posting.
- Relocation benefits are not available for this opportunity.
Tech Stack
OpenCVPyTorch
Categories
About Roche
Roche is a Swiss healthcare company that develops and sells prescription pharmaceuticals and in vitro diagnostic systems for hospitals, laboratories, and clinicians. Its business spans two core divisions—Pharmaceuticals and Diagnostics—with strengths in oncology, immunology, neuroscience, infectious diseases, and diabetes care. Founded in 1896 and headquartered in Basel, Switzerland, Roche also advances personalized healthcare by pairing therapies with companion diagnostics.