Job Description

• Contribute to crafting and executing on the client-wide research and development roadmap
• Develop, build, and apply modeling approaches—including adapting and post-training biological frontier models—for tasks in biological security and the design of precision biologics
• Develop, build, and apply evaluation frameworks to rigorously assess model performance on key tasks
• Interrogate models to understand their biological learnings, capabilities, and limitations
• Collaborate closely with computational biologists to understand the signal and artifacts inherent in biological data, and to create and curate datasets
• Collaborate closely with software engineers to build, deploy, and scale model training and evaluation infrastructure
• Visualize and communicate results within Client and externally
• Work with customers and external collaborators to understand their needs and be an effective representative of Client
• Embrace learning about areas technical and non-technical across Client
• Stay up-to-date on the state-of-the-art methods at the intersection of AI and biology

• Experience with data-centric development and evaluation of ML models
• Highly proficient in Python and deep learning libraries (e.g., PyTorch)
• Experience with cloud computing platforms (e.g., AWS) and distributed systems
• Strong understanding of probability and statistics

• Proven ability to design, implement, and evaluate new ideas in ML
• Experience with pre- or post-training language models, developing reasoning agents, and/or time series modeling
• Experience working with biological data or other types of noisy and heterogenous datasets
• Experience with ML-centric bioinformatics and structure-based tools (e.g., AlphaFold)
• Experience building data pipelining and training infrastructure
• Contributions to open-source projects