Co-Op, Autonomous SEM

Your Impact at LILA

Lila Sciences is seeking a Co-Op, Autonomous SEM to join the Materials Science team within the Autonomous Science Platform. This co-op will contribute to autonomous SEM workflows that make characterization more consistent, high-throughput, and less dependent on manual operators. The work sits at the intersection of materials characterization, image analysis, and autonomous laboratory workflows. The co-op will support SEM-based imaging workflows that help instruments identify useful regions, evaluate image quality, adjust acquisition conditions, and generate datasets suitable for downstream analysis and ML training.

This is a hands-on opportunity for a student interested in building practical autonomy for scientific instruments—turning SEM from a manually driven characterization tool into a system that can navigate samples, make acquisition decisions, and produce richer datasets for materials discovery.

What You'll Be Building

• Support development of autonomous SEM workflow using vendor APIs
• Test navigation logic for locating particles, surfaces, and regions of interest.
• Evaluate image quality using criteria such as focus, contrast, feature visibility, and sampling value.
• Support experiments that connect imaging decisions to downstream analysis and ML training needs.
• Document acquisition behavior, edge cases, and failure modes across sample types.
• Collaborate with ML scientists, experimental scientists, and software partners on instrument-control requirements.
• Help define practical guardrails for autonomous SEM operation, including when to capture, reposition, zoom, or adjust parameters.

What You'll Need to Succeed

• Currently pursuing a PhD or have completed a PhD in Materials Science, Chemistry, Chemical Engineering, Physics, Applied Physics, Computer Science, or a related technical field.
• Experience building automated scientific or laboratory workflows using Python.
• Deep understanding of electron optics, electron-beam interaction with matter, column alignments, stigmation correction, and source dynamics
• Familiarity with MCP servers, LLM-enabled workflows, or agentic control of scientific instruments.
• Experience with closed-loop learning, active learning, Bayesian optimization, or reward-driven experimental workflows.
• Ability to translate expert instrument operations into clear, testable, and well-documented workflow logic.

Bonus Points For

• Experience with autonomous microscopy, self-driving labs, or agentic scientific workflows.
• Experience with data analysis for scientific images, spectra, or microscopy datasets.
• Experience with image segmentation, particle finding, feature detection, or morphology analysis.
• Interest in building practical autonomy for scientific instruments across real sample types and workflows.