Scientist I Machine Learning for Generative Shape Modeling

$89,100 - $110,200/Yr

Allen Institute - Seattle, WA

Full-time - Entry Level

Remote - Seattle, WA

Ambulatory Health Care Services

About the position

The Scientist I - Machine Learning for Generative Shape Modeling position at the Allen Institute for Cell Science is a pivotal role aimed at advancing the understanding of cellular behavior through innovative computational methods. The mission of the Allen Institute is to create multi-scale visual models that elucidate cell organization, dynamics, and activities. This position is part of a team-oriented approach that emphasizes collaboration and the integration of diverse perspectives to drive high-quality scientific research. The Computational Cell Science team is focused on developing scalable, quantitative image-based analysis frameworks that can effectively analyze cell organization, activity, and function. In this role, the selected candidate will be responsible for developing machine learning workflows specifically designed for generative shape modeling from single-cell image data. This involves creating and implementing scalable and reproducible machine learning pipelines for 3D shape quantification based on data obtained from microscopy image-based assays. The candidate will be expected to systematically iterate on new models, maintain and improve existing machine learning pipelines, and collaborate closely with other teams within the institute to scale up analysis protocols into high-throughput computational pipelines. Ensuring the seamless integration and sharing of resources and data across teams, maintaining rigorous quality control standards, and adhering to standard operating procedures (SOPs) and good laboratory practices (GLPs) are also critical components of this position. The role requires meticulous record-keeping and coordination with other scientists to manage complex experiments effectively. The candidate will also be tasked with preparing written summaries and presenting their activities both internally and publicly. The Allen Institute is committed to fostering a diverse and inclusive work environment, encouraging individuals from all backgrounds to apply for this role, as diverse voices and experiences are seen as essential to producing high-quality science.

Responsibilities

• Develop and implement scalable and reproducible machine learning pipelines for 3D shape quantification on data from microscopy image-based assays.
• Systematically, efficiently and reproducibly iterate on new models.
• Maintain and improve existing machine learning pipelines.
• Work closely with other teams in the institute to scale-up analysis protocols into a high throughput computational pipeline.
• Ensure seamless integration and sharing of resources and data across teams.
• Maintain rigorous quality control standards.
• Maintain meticulous records and work closely with other scientists to coordinate complex experiments.
• Adhere to SOPs, GLPs and regulatory requirements.
• Prepare written summaries and present activities internally and publicly.

Requirements

• Ph.D. in Computational Physics, Applied Physics, Applied Mathematics, Computer Science, Biological Science (e.g. Cell Biology, Biophysics, Bioengineering), or related science or engineering field; OR equivalent combination of degree and experience.
• Extensive knowledge in machine learning and hands-on experience in developing and implementing deep learning algorithms and generative models like VAEs, GANs, autoregressive models and transformers.
• Experience with image-based biology assays; some experimental and/or microscopy image analysis experience would be an advantage.
• Experience with different data representations like images, point clouds, meshes; some computational geometry experience would be an advantage.
• Experience with developing or contributing to open-source tools/packages.
• Experience utilizing software engineering practices such as version management, build management and testing; Experience with MLOps tools like MLflow, prefect would be an advantage.
• Careful attention to detail.
• Excellent interpersonal skills.
• Experience working in a multi-disciplinary environment in academic or industrial settings.
• Ability to work both independently and in a collaborative, multi-disciplinary environment.

Nice-to-haves

• Experience with MLOps tools like MLflow, prefect would be an advantage.
• Some computational geometry experience would be an advantage.

Benefits

• Health insurance
• Dental insurance
• 401(k)
• Paid time off
• Vision insurance
• Life insurance