Machine Learning Scientist - Small Molecule Binding

🚀 About Lila

Lila Sciences is the world’s first scientific superintelligence platform and autonomous lab for life, chemistry, and materials science.  We are pioneering a new age of boundless discovery by building the capabilities to apply AI to every aspect of the scientific method.  We are introducing scientific superintelligence to solve humankind's greatest challenges, enabling scientists to bring forth solutions in human health, climate, and sustainability at a pace and scale never experienced before. Learn more about this mission at  www.lila.ai

If this sounds like an environment you’d love to work in, even if you only have some of the experience listed below, we encourage you to apply.

🌟 Your Impact at Lila

Join our Physical Science division to invent and deploy learning-based approaches for small molecule docking, pose prediction, and binding affinity estimation. You’ll partner with computational chemists, structural biologists, and assay teams to build physics-informed, geometry-aware models that reason over proteins, pockets, ligands, and solvent and co-factors. Your work will power high-precision virtual screening, generative design, and closed-loop discovery for real drug targets and chemical modalities.

This role complements our ligand-based optimization team by emphasizing structure- and physics-informed docking, pose prediction, and affinity modeling—even when assay data are sparse or noisy—ultimately enabling pocket-guided design, high-precision virtual screening, and faster DMTA cycles with improved candidate quality.

🛠️ What You'll Be Building

• Structure-aware docking and scoring models: Develop SE(3)-equivariant, graph, point cloud, and 3D grid models for protein–ligand complexes that capture interactions, sterics, electrostatics, and symmetry; handle metalloproteins, covalent chemistries, and allosteric pockets.
• Pose prediction and affinity estimation: Build models for binding pose ranking and ΔG prediction with physics-informed objectives, multi-instance/ensemble learning, and integration of energy terms and constraints.
• Generative design and search: Create and deploy diffusion/flow/RL-based methods to generate poses and ligands conditioned on pockets, fragment-growing/linking, constrained optimization for potency, selectivity, and developability.
• Protein flexibility and induced fit: Incorporate receptor ensembles (MD, NMA), side-chain sampling, pocket conformer selection, and induced-fit strategies; couple ML with MD, MM/GBSA, and free-energy workflows when needed.
• Chemistry realism: Model protonation/tautomer states, stereochemistry, water networks and displaceable waters, metal coordination, and covalent warheads; robust enumeration and state handling in training/inference.
• Screening under uncertainty: Calibrated uncertainties and acquisition strategies to guide synthesis, assays, and simulations; triage ultra-large libraries and reduce false positives.
• Data and evaluation foundations: Curate high-quality, leakage-controlled datasets (e.g., PDBbind, CrossDocked/CASF, BindingDB, ChEMBL); define rigorous benchmarks for cross-target generalization and prospective validation.
• Real-world deployment: Build scalable, GPU-accelerated virtual screening services and APIs; integrate with cheminformatics and simulation stacks; monitor drift and maintain reliability in production.
• Cross-functional partnership: Work closely with R&D leadership, product, and lab automation to translate targets into ML workflows and close the loop with experimental validation.

🧰 What You’ll Need to Succeed

• Strong proficiency in Python and modern deep learning (PyTorch, JAX, or TensorFlow), including training at scale, experiment tracking, and deployment.
• Expertise in representation learning and geometric deep learning for 3D molecular data (equivariance, point/graph methods) applied to protein–ligand systems.
• Solid grasp of protein–ligand interactions, docking/scoring, and the thermodynamics/statistical mechanics underlying binding.
• Experience with cheminformatics and structural biology tooling: RDKit, PyTorch Geometric/DGL, e3nn, OpenMM/MDTraj/ParmEd, OpenFF, BioPython; familiarity with docking/physics engines (e.g., AutoDock Vina/Smina/Gnina, Rosetta/PyRosetta, MM/GBSA, FEP/TI).
• Ability to design rigorous evaluations (pose RMSD, success@k, CASF metrics, cross-target/OOD tests, uncertainty calibration) and to build leakage-safe datasets and splits.
• End-to-end ownership of ML pipelines: data curation, feature/graph generation, training, serving, and monitoring in cloud/HPC environments.
• Strong self-starter with excellent attention to detail and clear, collaborative communication.
• Demonstrated industry experience or academic achievement.

✨ Bonus Points For

• PhD in Computer Science, Computational Chemistry, Chemistry, Biophysics, or related field with a strong publication record in ML (NeurIPS, ICML, ICLR) and/or top-level computational chemistry/structural biology venues.
• Prior work on differentiable docking/scoring, SE(3)-equivariant message passing, or structure-based foundation models for pockets/proteins/complexes.
• Experience integrating ML with MD and free-energy methods (alchemical FEP, TI, PMF) and using receptor ensembles for flexible docking.
• Familiarity with protein structure prediction and modeling (AlphaFold2/ESMFold, homology modeling), pocket detection, and cryo-EM/X-ray refinement.
• Experience modeling waters (GIST/WaterMap), protonation/pKa workflows, metal coordination, covalent docking, and allosteric/cryptic sites.
• Building ultra-large virtual screening pipelines (billions-scale) with efficient indexing/embedding search, cloud/HPC scaling, and robust MLOps.

🌈 We’re All In

Lila Sciences is committed to equal employment opportunity regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity or Veteran status.

🤝 A Note to Agencies

Lila Sciences does not accept unsolicited resumes from any source other than candidates. The submission of unsolicited resumes by recruitment or staffing agencies to Lila Sciences or its employees is strictly prohibited unless contacted directly by Lila Science’s internal Talent Acquisition team. Any resume submitted by an agency in the absence of a signed agreement will automatically become the property of Lila Sciences, and Lila Sciences will not owe any referral or other fees with respect thereto.