The Opportunity

To accomplish our mission, we need to develop a dynamic, integrated view of biological systems in health and disease. As an extension of the Chan Zuckerberg Initiative's Imaging Program, the CZ Imaging Institute aims to push forward the development and application of new imaging technologies over the next 15 years to obtain deep insights into the architecture of complex biological systems at the molecular level.

The CZ Imaging Institute will create breakthrough technologies — hardware, software, biological probes, data, and platforms — that will be made available to the scientific community and adopted worldwide through a combination of direct access to the institute, open sharing of advances, and commercial partnerships. Researchers will collaboratively develop breakthrough biological imaging systems centered around grand challenges that push the boundaries of what we can see and measure.

The Chan Zuckerberg Imaging Institute is seeking an experienced AI/Machine Learning Research Scientistto support the New Algorithms for Cryo ET Team Leader to create innovative algorithms that address specific challenges in computer vision, interpreting and enhancing Cryo-EM data while developing a robust framework to guard against over-interpretation. Will work closely with teams of scientists, computational biologists, engineers within the CZ Imaging Institute and will collaborate with CZI grantees, with CZ institutes, and other external labs and organizations.

What You'll Do

• Develop and apply state-of-the-art methods in artificial intelligence and machine learning to solve important problems in biological imaging such as increasing contrast of images, resolution of images, and throughput of data.
• Collaborate with the computer vision team at CZI, focusing on developing methods that automate the interpretation of cryo-ET data.
• Work as part of a team to leverage large open-source biological datasets to create models that transform our understanding of cells.
• Bridge information from alternate modalities, like protein language models, to decode Cryo-EM data.
• Leverage constraints derived from foundation models, which take into account both the cellular context and the conformational landscape of individual protein complexes, available in tomography data.
• Work closely with researchers and scientists developing novel approaches in cryoEM.

What You'll Bring

Essential –

• PhD in Software Engineering or Machine Learning
• 1-2 years experience working with AI/ML tools and Algorithms
• Have experience in representation learning techniques, including expertise in unsupervised and self-supervised learning methods, dimensionality reduction, and feature extraction and the ability to create meaningful and high-quality data representations that facilitate downstream tasks
• A track record of creating innovative algorithms that address specific challenges in computer vision, inverse problems, and related fields, including a proven ability to adapt and combine existing techniques or develop novel approaches to solve complex problems
• Hands-on experience with deep learning frameworks (e.g. PyTorch, TensorFlow, JAX) and a solid understanding of neural network architectures for tasks like image recognition, segmentation, and generative modeling
• History of research and innovation in machine learning and computer vision, demonstrated through publications, conference presentations, or project outcomes with an ability to stay updated with the latest advancements in the field
• Effective teamwork and communication skills, including the ability to work collaboratively in cross-functional teams, present complex technical concepts to non-technical stakeholders, and contribute to the growth and mentorship of junior team members
• Hands-on experience with Graph Neural Networks (GNNs) and a strong grasp of graph-based representation learning including the ability to design, implement, and optimize GNNs for various applications, showcasing the ability to capture intricate relationships within complex data structures

Nice to have –

• Proficiency in multimodal learning, including experience in developing models that effectively fuse information from different data modalities (such as text, images, graphs, etc.) to create a comprehensive understanding of the underlying data.
• Possessing a deep understanding of advanced machine learning techniques, including proficiency in training and fine-tuning large language models with an ability to showcase expertise in working with various architectures and methodologies to harness the power of language models for various tasks.
• Experience applying machine learning expertise to the domain of protein folding with a demonstrated experience in developing models that contribute to advancements in protein folding research

Compensation

• Leveling = 6
• Salary Rage = $131,000 - $180,400

New hires are typically hired into the lower portion of the range, enabling employee growth in the range over time. To determine starting pay, we consider multiple job-related factors including a candidate's skills, education and experience, market demand, business needs, and internal parity. We may also adjust this range in the future based on market data. Your recruiter can share more about the specific pay range during the hiring process.