Scientist II/Senior Scientist, Computational Biophysics

About the role

This position centers on architecting and validating the computational biophysics foundations that allow autonomous scientific agents to execute reliable drug discovery experiments. You will design the frameworks that translate physical simulations into structured inputs and guarded outputs for AI-driven chemistry agents. A central ownership is ensuring that molecular simulation theory, binding free-energy methods, and statistical mechanics are implemented with traceable and reproducible standards. You will own the validation criteria that determine when a simulation is sufficiently accurate to inform chemical decision-making. This role requires you to act as the scientific authority on what biophysical experiments and computational methods can be safely delegated to automated pipelines. You will define the guardrails that prevent mis-specified models, incorrect force fields, or unstable setups from propagating through agent workflows. Ultimately, you own the bridge between rigorous computational biophysics and scalable, agentic discovery platforms that non-specialists can safely invoke.

Key facts

What you'll do

• Architect and maintain the scientific validation frameworks that enable autonomous agents to correctly use biophysics tools in drug discovery pipelines.
• Design molecular simulation workflows for binding free-energy calculations, specifying inputs, assumptions, and decision-grade outputs for automated use.
• Establish standards for system setup, force field selection, molecular parameterization, equilibration, sampling, analysis, and quality control across target classes.
• Build validated RBFE and ABFE workflows that include automated checks for chemical-series compatibility, ligand-pose consistency, and thermodynamic cycle integrity.
• Partner with research engineers to translate scientific protocols into reliable tools, APIs, and guardrails that can be safely invoked by large language models.
• Create and maintain validation benchmarks and acceptance criteria that define when molecular dynamics, binding free-energy, and related workflows meet scientific and operational standards.
• Evaluate and integrate open-source molecular simulation and free-energy tools, with a focus on OpenMM, OpenFE, and related packages used in modern discovery stacks.
• Identify, document, and mitigate failure modes across simulation setup, parameterization, sampling, analysis, and interpretation to reduce downstream risk.
• Advise cross-functional teams on the appropriate application of biophysics workflows, the evidence they support, and their practical limitations in discovery contexts.
• Collaborate closely with ML and infrastructure teams to ensure that methods and parameters are refined with new experimental data and remain scalable across large GPU fleets.
• Implement traceability and reproducibility practices that link simulation configurations, code versions, and data lineage to decision records.
• Translate complex scientific methods into clear tooling requirements, review standards, and operational workflows that non-specialist engineers can adopt.
• Support the integration of validated biophysics tools into automated experimental feedback loops where ML models propose molecules and simulations evaluate them.
• Champion continuous improvement of validation metrics, benchmarking datasets, and failure-mode catalogs to strengthen agent reliability over time.
• Represent Lila Sciences in the computational biophysics community by contributing to open-source ecosystems around simulation and free-energy methods.

Requirements

• Hold a PhD or equivalent experience in computational biophysics, computational chemistry, chemical physics, biophysics, physics, or chemistry.
• Demonstrate hands-on experience with molecular dynamics simulation, including MM/GBSA and free-energy perturbation methods for RBFE and ABFE.
• Possess practical experience with open-source molecular simulation and free-energy packages, with strong preference for OpenMM and OpenFE.
• Show a deep understanding of how to parameterize molecular systems and validate that simulation setups are scientifically sound.
• Have a record of modeling protein-ligand binding and interpreting results to support scientific decision-making.
• Exhibit the ability to reason from biophysical first principles while building practical workflows usable by cross-functional teams.
• Be skilled at translating complex scientific methods into tooling requirements, review standards, and operational workflows.
• Communicate effectively with computational scientists, ML researchers, research engineers, and drug discovery teams in a collaborative environment.

Nice to have

• Experience contributing to community simulation and free-energy tool ecosystems is a bonus.

Practical notes

Please What you'll do