Computational Scientist I/ii, Soft Matter Formulations, Solids and Melts

Lila Sciences Lila Sciences · AI Frontier · Alewife, Cambridge, MA · Physical Sciences AI

This role involves developing machine learning models and active learning workflows to accelerate discovery in soft matter and material systems, connecting formulation choices to performance through structure-property models.

What you'd actually do

  1. Develop machine learning models for polymers, elastomers, gels, hot-melts, adhesives, composites, powders, films, semi-solids, and crystalline or amorphous solids.
  2. Define modeling targets for mechanical performance, thermal transitions and processing windows, and processing-sensitive material responses.
  3. Build representations that connect formulation variables, processing history, morphology, crystallinity, cross-link density, cure kinetics, and end-use properties.
  4. Develop structure-property models for solid and viscoelastic materials using experimental, simulation, rheological, thermal, mechanical, and formulation datasets.
  5. Build active learning workflows that prioritize formulation experiments in line with physical formulation workcell throughput and lab constraints.

Skills

Required

  • Experience applying machine learning to scientific, materials-focused, polymer, soft matter, or formulation problems.
  • Domain expertise in polymer science, elastomers, gels, adhesives, composites, rheology, solid materials, complex fluids, or related fields.
  • Familiarity with mechanical, thermal, morphological, or processing-sensitive material properties
  • Strong Python skills and experience with modern ML frameworks.
  • Experience training, evaluating, and improving models using experimental, simulation, or scientific datasets.
  • Ability to use simulations, theory, descriptors, or mechanistic understanding to inform modeling choices for polymer and soft material systems.
  • Strong communication skills with experimental, computational, and cross-functional collaborators.
  • PhD in chemical engineering, materials science, physics, applied mathematics, computational science, or a related field, or a master’s degree with equivalent relevant experience.

Nice to have

  • Experience working with experimental data from polymers, elastomers, gels, hot-melts, adhesives, composites, powders, films, semi-solids, or solid formulations.
  • Experience modeling structure-property relationships for solid, semi-solid, or viscoelastic materials.
  • Familiarity with cure- or processing-aware representations for formulation, thermal, mechanical, or rheological datasets.
  • Experience incorporating molecular or polymer descriptors, simulation constraints, theoretical models, or mechanistic priors into ML workflows.
  • Background in active learning systems that close the loop between models and high-throughput physical experimentation.
  • Hands-on experimental or computational experience in polymer, adhesive, gel, composite, or soft material formulation domains.

What the JD emphasized

  • apply machine learning methods
  • develop structure-property models
  • build active learning workflows

Other signals

  • applying machine learning methods to connect formulation choices, processing history, structure, morphology, and end-use performance
  • develop structure-property models for solid and viscoelastic materials
  • build active learning workflows tied to the throughput of the physical formulation workcell