Machine Learning Research Scientist, SandboxAQ

• Domain Foundation: Ph.D. in Computational Biology, Biophysics, Computer Science, Computational Chemistry, or a related field, with a research focus on protein structure prediction, co-folding, or closely related areas
• Hands-On Co-Folding Experience: Direct experience with protein structure prediction or protein-ligand co-folding methods (e.g., AlphaFold2/3, RoseTTAFold, Chai-1, Boltz, or comparable systems), developed through graduate or postdoctoral research
• ML Model Development: Experience developing, training, and validating deep learning models, including familiarity with architectures relevant to structural biology (e.g., transformers, equivariant neural networks, diffusion models)
• Programming Proficiency: Strong proficiency in Python and modern ML frameworks (PyTorch and/or JAX)
• Scientific Rigor: Demonstrated ability to design controlled experiments, interpret results critically, and iterate effectively on model development
• Communication and Collaboration: Strong written and verbal communication skills; ability to work collaboratively in a fast-paced, multidisciplinary research environment

Desirable

• The AI Sim R&D team creates leading edge ML and physics-based models ("LQMs") to advance drug and materials discovery
• We are a flexible, creative, and impact driven team of multidisciplinary scientists and engineers, whose products dramatically accelerate the creation of molecules and medicines
• As a Research Scientist focusing on Co-Folding & Affinity, you will contribute to building SandboxAQ's next generation of structure prediction and binding affinity models
• Working alongside a high-performing team of scientists and engineers, you will help advance the state-of-the-art in protein-ligand co-folding, translating cutting-edge research into scalable workflows that power our drug discovery software
• This is an opportunity to do frontier science with real-world impact — developing models that redefine what's possible in computational drug discovery
• Develop and Iterate on Co-Folding Models: Implement, experiment with, and refine deep learning models for protein-ligand co-folding and structure prediction, building on the latest research from the field
• Drive Rigorous Benchmarking: Design and execute systematic evaluation pipelines to measure model performance against state-of-the-art methods and internal benchmarks
• Contribute to Research-to-Product Pipelines: Collaborate with senior scientists and engineers to integrate validated models into production-ready drug discovery workflows
• Apply Data-Driven Methods: Employ computational and data analysis techniques to generate insights from structural and sequence datasets, informing model development decisions
• Communicate Findings: Present research progress through internal scientific talks, technical write-ups, and contributions to peer-reviewed publications
• Collaborate Across Teams: Work closely with multidisciplinary teams — including ML engineers, structural biologists, and software engineers — to prototype and scale impactful solutions