Machine Learning Engineer II / Senior Machine Learning Engineer I, Physical Sciences

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

Machine Learning Engineer role focused on building and operating end-to-end, scalable ML workflows for scientific use cases in materials, chemistry, and physical sciences. The role involves designing, implementing, and maintaining ML pipelines, productionizing models and services, and collaborating with domain scientists and platform engineers to translate research insights into scalable systems. Experience with model deployment in production, including LLMs and multimodal models, is required.

What you'd actually do

  1. Design, implement, and maintain end‑to‑end ML pipelines (data ingestion, feature engineering, training, evaluation, deployment, monitoring).
  2. Productionize models and services with robust testing, observability, and documentation in collaboration with cross-functional software teams and build CI/CD workflows and automated evaluations to ensure safe, frequent releases.
  3. Collaborate with domain scientists and platform engineers to translate research insights into performant, scalable systems.
  4. Contribute to technical design reviews, coding standards, and mentoring of best practices.

Skills

Required

  • Python software engineering fundamentals
  • machine learning frameworks (e.g., PyTorch, Huggingface, etc.)
  • deploying ML services to production in cloud-based infrastructure (FastAPI/GRPC, containers, orchestration, cloud infra)
  • model deployment in production systems (LLMs, multimodal models, databases, RAG)
  • debugging and profiling skills
  • Clear communication and collaboration in cross‑functional settings

Nice to have

  • scientific or engineering domains (materials, chemistry, physics) and related data formats/benchmarks
  • GPU optimization experience (CUDA, Triton, compilation, distributed training)
  • Prior contributions to open‑source ML or scientific software
  • workflow orchestration, data provenance, or large‑scale compute environments

What the JD emphasized

  • end-to-end ML pipelines
  • productionize models and services
  • model deployment in production systems

Other signals

  • end-to-end ML pipelines
  • productionize models and services
  • deploy ML services to production
  • model deployment in production systems