Machine Learning Scientist, Generative Molecular Design – Material Discovery

This is a fully remote position, open to applicants in United States.

📋 Description

• Design and advance SandboxAQ's premier generative platform for material discovery.

• Implement, enhance, and optimize cutting-edge generative architectures.

• Navigate complex high-dimensional chemical spaces and address multi-objective optimization challenges.

• Integrate learned surrogates and simulation-derived oracles as indicators of performance.

• Craft reward formulations that balance competing objectives, such as performance, environmental impact, and safety constraints.

• Develop and deploy computational methods and workflows to generate and assess hypotheses.

• Steer design choices and shape project trajectories.

• Identify pertinent data sources and take ownership of data ingestion, quality control, and validation processes.

• Collaborate closely with ML researchers to implement advanced generative architectures.

• Work alongside R&D, discovery, engineering, and product teams.

• Transform models into production capabilities that discovery scientists can utilize.

⛳️ Requirements

• A minimum of 4+ years of experience delivering state-of-the-art generative or applied ML solutions for material discovery in an industry or comparable research environment.

• Practical experience in building and deploying modern generative or applied ML solutions for material discovery.

• Proficiency in writing clean, maintainable, and robust production-ready code for large-scale applications.

• Capability to build scalable ML systems using PyTorch, TensorFlow, or JAX.

• Familiarity with cloud-based platforms such as Google Cloud Platform or AWS for computing, data processing, and storage.

• PhD or MSc in Computational Chemistry, Physics, Computer Science, or a related field.

• Exceptional communication skills with a proven track record of fostering successful internal collaboration.

• Experience in collaboratively developing ML code utilizing modern software engineering best practices.

• Experience with active learning and Bayesian optimization is highly preferred.

• Background in physics simulations like DFT or molecular dynamics, or experimental lab-in-the-loop assays, is highly preferred.

• Experience in guiding generative modeling for material discovery is highly preferred.

• Familiarity with diffusion, flow-matching, GFlowNets, LLMs, agent-based methods, guided diffusion, sequential Monte Carlo, LoRA, or reinforcement learning is highly preferred.

• Knowledge of agentic and LLM approaches to material discovery is highly preferred.

🏝️ Benefits

• Equity.

• Performance-based incentives.

• Comprehensive health, dental, and vision insurance.

• 401(k) plan with company matching.

• Generous parental leave.

• Flexible hybrid work arrangements.

• Generous paid time off (PTO).

• Direct exposure to CHIPS Act-funded initiatives.

• Mentorship opportunities.

• Dedicated budgets for learning and development.

• Opportunities for professional advancement.