The Team
Four people with specific, non-interchangeable expertise in electrochemical engineering, industrial time-series ML, dQ/dV analysis, and gigafactory operations. No generalists — everyone on this team has worked inside the exact problem Moldpathio is solving.
Spent several years leading process quality engineering at a battery cell manufacturer in Nevada before founding Moldpathio. Her direct observation of formation-stage SPC blind spots — defects that only surfaced at final QC, long after the process had already moved on — was the founding thesis. M.S. Electrochemical Engineering, University of Utah.
Built and deployed time-series anomaly detection systems for industrial process monitoring before co-founding Moldpathio. Brings the inference architecture — the sliding-window dQ/dV decomposition, TCN scoring loop, and per-site calibration pipeline that makes the 5-minute scoring cadence possible on a single edge node. B.S. Computer Engineering; ML systems background in predictive maintenance for continuous manufacturing processes.
PhD candidate in computational electrochemistry at the University of Utah, where her dissertation focuses on SEI formation dynamics under non-isothermal conditions. Leads Moldpathio's feature engineering work — translating electrochemical theory into ML-legible signal features. Designed the defect taxonomy underlying the six alert classes the platform currently supports.
Came from manufacturing operations at automotive Tier-1 suppliers, where he managed process ramp programs for new production lines. At Moldpathio, he owns every pilot deployment from edge connector install through model calibration to steady-state operations handoff. The customers team members quote in case studies are his pilots. Understands both the language of process engineering and the constraints of a gigafactory operations schedule.
We are a focused team at an early stage. If you have specific experience in electrochemical engineering, time-series ML for manufacturing, or gigafactory operations — and the problem of formation-stage defect detection interests you — write to us. We hire for domain depth, not credential breadth.
[email protected]