Machine Learning Scientist – Molecular Force Fields

SandboxAQ

. Design and run high-throughput quantum chemistry and molecular dynamics campaigns producing training data for PFAS chemical space .

Posted 10/26/2026full-timeRemote • United StatesMid-LevelSenior💰 $134,400 - $252,000 per yearWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Expertise in developing and deploying machine learned interatomic potentials and conducting high-throughput quantum chemistry simulations. Proficient in building automated workflows and curating scientific datasets while ensuring reproducibility and validation discipline.

Highest-signal resume keywords

PhD In Computational ChemistryMachine Learned Interatomic PotentialsQuantum Chemistry Codes (VASP, Quantum ESPRESSO)Python Programming SkillsMachine Learning Frameworks (PyTorch, JAX)

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Machine Learned Force FieldsMolecular Dynamics SimulationDescriptor-Based Potential ArchitecturesProduction Prediction CampaignsError Budget AnalysisAutomated Computational WorkflowsClassical Molecular DynamicsForce Field ParameterisationConverged Physical Observables ExtractionGPU Cluster Training

Soft Skills

Judgment In Setting Theory LevelsCollaboration With Scientific Teams

Tools & Technologies

LAMMPSGROMACSOpenMMASESandboxAQ Tooling

Industry Keywords

PFAS Chemical SpaceHigh-Performance ComputingCloud ComputingScientific Datasets CurationReproducibility

Tech Stack

Tools & technologies

CloudPythonPyTorch

About the role

Key responsibilities & impact

Design and run high-throughput quantum chemistry and molecular dynamics campaigns producing training data for PFAS chemical space
Own reproducible infrastructure for physics-based simulation data generation
Determine level of theory, basis set, convergence criteria, sampling strategy, and error budget
Build automated, restartable, provenance-tracked workflows across cloud and HPC
Curate and version internally generated and public molecular and materials datasets
Select architectures, train, validate, and deploy machine-learned interatomic potentials for PFAS substitution
Fine-tune universal or foundation potentials, or train models from scratch when appropriate
Build and validate MLFF-driven protocols for protonation/deprotonation energetics, ion diffusivity, viscosity, and thermal conductivity
Extend potentials to resin-phase polymerisation and plasma surface etch chemistry
Execute production prediction campaigns across candidate libraries
Deploy trained models, datasets, and reproducible training pipelines in SandboxAQ tooling
Collaborate with generative modeling, molecular property ML, simulation, and materials science teams
Report error bars, hold-out performance, and domain-of-validity limits for predicted properties
Escalate when models are asked questions beyond their capability

Requirements

What you’ll need

PhD in Computational Chemistry, Materials Science, Physics, Machine Learning, or a related field, with deep specialization in machine learned interatomic potentials or atomistic simulation
Demonstrated depth developing and deploying machine learned force fields, including at least one potential trained, validated, and put into the hands of other people who rely on it
Hands-on experience generating first principles training data at throughput with production quantum chemistry and electronic structure codes such as VASP, Quantum ESPRESSO, CP2K, ORCA, Psi4, or Gaussian
Judgment to set and defend level of theory, basis set, and convergence criteria
Fluency with descriptor-based, modern equivariant, and message-passing potential architectures such as MACE, NequIP, Allegro, TensorNet, DeePMD, ANI-class, or FairChem/UMA-family models
Strong programming skills in Python
Hands-on experience with modern ML frameworks such as PyTorch and JAX
Experience training at scale on GPU clusters
Practical experience coupling learned potentials into production molecular dynamics engines such as LAMMPS, GROMACS, OpenMM, or ASE
Experience extracting converged physical observables from molecular dynamics trajectories
Hands-on experience with classical molecular dynamics, force field selection and parameterisation for organic and condensed phase systems
Experience with OPLS-AA, AMBER/GAFF, or ReaxFF for reactive chemistry
Experience with system construction and equilibration, ensemble and thermostat/barostat choice
Experience building automated computational workflows and curating large scientific datasets with provenance, versioning, and reproducibility
Demonstrated validation discipline, including characterising model error and out-of-domain failure
Experience working in cloud and high performance computing environments
Candidates cannot participate in foreign talent programs sponsored by China, Russia, Iran, or North Korea due to federal grant requirements

Benefits

Comp & perks

Competitive base salary commensurate with experience
Equity
Performance-based incentives
Comprehensive health, dental, and vision insurance
401(k) with company match
Generous parental leave
Flexible hybrid work arrangements
Generous PTO
Direct exposure to CHIPS Act-funded programs
Mentorship
Dedicated learning budgets
Reasonable accommodations for individuals with disabilities in job application procedures