About the role
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.
