True battery intelligence requires understanding the full lifecycle. I combine deep expertise in cell selection, pack design, production and application with advanced Machine Learning. I don't just "play with" data; I understand the whole pictures behind the data.
Chemistry, suppliers, and QC baselines locked in.
Thermal paths, harnessing, and redundancy decisions.
Batch genealogy links cell, pack, and line conditions.
Physics-informed AI flags early failure signatures.
Why pure Data Science often fails in Battery Engineering.
Standard data scientists treat physical anomalies as statistical noise. They "clean" the data to fit a curve, inadvertently stripping away the subtle signals of lithium plating or thermal degradation because they lack the physical context.
I know the physics of the cell from development to race day. I preserve "outliers" that represent real physical limits. I use AI not just to fit curves, but to uncover the complex, non-linear relationships between manufacturing inputs and track performance.
Traditional development isolates cell data from pack manufacturing. I build systems that preserve information integrity across the entire value chain, ensuring no data is lost in handoffs.
I developed a proprietary system linking Cell EOL data to Pack Manufacturing. This creates a permanent digital passport for every battery pack.
When a track issue arises, we don't guess. We trace. My systems allow instant drill-down from vehicle telemetry back to specific cell batch manufacturing conditions.
