Senior Engineer, Machine Learning
Job description
About the Role
This position centers on the design and delivery of machine learning systems for biological imaging. You will lead the development of models within a product-focused scientific environment, working alongside instrumentation and biology teams. Success requires a commitment to scientific rigor, transparent communication, and objective decision-making. The role is based onsite at our headquarters in San Diego.
Key Facts
-
Location: USA
-
Engagement: Onsite.
-
Compensation: $180,000 to $220,000 USD annually.
Responsibilities
You will architect and optimize neural networks tailored for cellular image analysis, including CNNs, Vision Transformers, and U-Net variants. A core focus is deploying these models to production settings, whether on cloud infrastructure like AWS or directly on imaging devices, where reliability and performance are paramount. You will handle the full lifecycle of machine learning pipelines, encompassing data ingestion, preprocessing, training, validation, and inference.
A significant portion of the work involves applying advanced computer vision methods to biological image data. This includes tasks such as segmentation, feature extraction, and automated quality scoring of multimodal images. You will analyze single-cell and multiomic datasets to provide biological context and use these insights to refine imaging models. Collaboration is integral; you will work closely with cross-functional partners to translate biological experimental needs into modeling goals.
Analyzing large-scale imaging datasets to identify performance bottlenecks and failure modes will guide future model and data strategies. You will also communicate complex technical findings, including performance metrics and trade-offs, to stakeholders through reports and presentations. Continuous learning is essential, as you will monitor new developments in your field to assess their value for our research pipeline.
Requirements
A Master's degree in a relevant discipline such as Computer Science, Bioinformatics, or Computational Biology is required, with five to seven years of related experience. Alternatively, a PhD can substitute for this experience. You must have a proven track record of deploying image analysis models in production, whether in cloud environments or on-device.
Practical expertise with core deep learning architectures is mandatory, specifically CNNs, Vision Transformers, and U-Net, including attention-based mechanisms. Prior experience with biological imaging data is highly valued. You must be proficient in Python and its data science stack, including PyTorch, OpenCV, Scikit-learn, NumPy, and Pandas.
Experience with cloud platforms, containerization tools like Docker, and GPU compute environments is required. Knowledge of model calibration and uncertainty quantification is a strong advantage. A solid foundation in statistics and quantitative analysis of model outputs is necessary. You must excel at solving complex problems and working across scientific and engineering boundaries.
Nice to Have
Experience with single-cell analysis and multiomic data workflows, along with knowledge of experimental design, is a plus but not required.
Skills & Tools
- Frameworks: PyTorch, torchvision, OpenCV, Scikit-learn, NumPy, Pandas
- Platforms: AWS, Docker
Practical Notes
Please confirm all details on the official application page before submitting your materials.
What you'll do
Element Biosciences builds tools that let researchers probe cellular activity with imaging and computation. Teams design assays, image analysis, and data models that translate raw readouts into biological insight.
You work at the interface of wet-lab experiments and software pipelines. Your contributions shape experiments, influence analysis, and connect measurement to biological questions.
About the Role
You will architect and deploy machine learning systems that transform biological imaging data into actionable scientific insight. This role demands ownership of the end-to-end model lifecycle, from initial experimental hypothesis to productionized inference on cloud platforms and embedded devices. You will act as a technical leader, ensuring that scientific rigor and objective decision-making guide every modeling choice. Success requires transparent communication with biologists and instrumentation engineers to align model behavior with wet-lab reality. You will operate within a product-focused environment where reliability, reproducibility, and measurable impact are paramount. The position is fully onsite at our headquarters in San Diego, requiring deep collaboration across multidisciplinary teams. Your work will directly influence the tools that researchers use to probe cellular activity and interpret complex biological readouts.
Key Facts
-
Location: San Diego, Headquarters.
-
Engagement: Onsite.
-
Compensation: $180,000 to $220,000 USD annually.
What you'll do
- Architect and optimize neural networks tailored for cellular image analysis, including CNNs, Vision Transformers, and U-Net variants.
- Deploy models to production settings, ensuring reliability and performance on cloud infrastructure like AWS and directly on imaging devices.
- Handle the full lifecycle of machine learning pipelines, encompassing data ingestion, preprocessing, training, validation, and inference.
- Apply advanced computer vision methods to biological image data, performing segmentation, feature extraction, and automated quality scoring of multimodal images.
- Analyze single-cell and multiomic datasets to provide biological context and refine imaging models based on these insights.
- Collaborate with cross-functional partners to translate biological experimental needs into concrete modeling goals and evaluation criteria.
- Analyze large-scale imaging datasets to identify performance bottlenecks and failure modes, guiding future model and data strategies.
- Communicate complex technical findings, including performance metrics and trade-offs, to stakeholders through reports and presentations.
- Monitor new developments in machine learning and computer vision to assess their value for our research pipeline and potential integration.
- Conduct rigorous experimentation to evaluate model behavior, ensuring that outputs are trustworthy and aligned with scientific objectives.
Requirements
- Hold a Master's degree in a relevant discipline such as Computer Science, Bioinformatics, or Computational Biology.
- Possess five to seven years of related professional experience, or a PhD to substitute for this requirement.
- Demonstrate a proven track record of deploying image analysis models in production, whether in cloud environments or on-device.
- Exhibit practical expertise with core deep learning architectures, specifically CNNs, Vision Transformers, and U-Net, including attention-based mechanisms.
- Show prior experience with biological imaging data, which is highly valued for success in this role.
- Be proficient in Python and its data science stack, including PyTorch, OpenCV, Scikit-learn, NumPy, and Pandas.
- Have experience with cloud platforms, containerization tools like Docker, and GPU compute environments.
- Possess knowledge of model calibration and uncertainty quantification, which is a strong advantage.
- Maintain a solid foundation in statistics and quantitative analysis of model outputs.
- Excel at solving complex problems and working across scientific and engineering boundaries.
- Commit to adhering to the highest standards of scientific integrity and data security in all modeling activities.
- Thrive in a fast-paced, product-driven environment where adaptability and ownership are essential.
Nice to Have
- Experience with single-cell analysis and multiomic data workflows, along with knowledge of experimental design.
Skills & Tools
- Frameworks: PyTorch, torchvision, OpenCV, Scikit-learn, NumPy, Pandas
- Platforms: AWS, Docker
Practical Notes
Please confirm all details on the official application page before submitting your materials.