Not tutorials. Not retrospectives with tidy conclusions. These are written after enough time has passed to know what actually mattered and what was just noise at the time.
Each one starts from a real problem — a model that kept breaking, a tool that kept growing, a decision that looked obvious in hindsight and wasn’t at all in the moment.
On simulation and modelling
-
Simulation and Modelling — Fifteen Years A walk through how my thinking on automotive, battery, thermal, and ageing modelling evolved — the approaches that broke and the ones that held.
-
The Battery Model That Kept Getting Bigger From RC circuits to multi-physics: how a battery model grows from an equivalent circuit into a multi-scale, multi-timescale problem — and why the answer ends up hybrid.
-
Debugging a Black Box vs Debugging Physics The real argument for physics-based models isn’t accuracy — it’s that they carry the signature of the physics inside them, which makes them debuggable.
-
Tools, Projects, Frameworks — and the Difference Why building a simulation framework is a fundamentally different job than building a tool or finishing a project — and how that distinction kept repeating itself across battery and AI work.
On AI and knowledge
-
What AI Can’t Inherit On the difference between knowledge and experience — and why a knowledge-graph RAG system can, and can’t, close that gap.
-
Exploring Local LLMs How running local AI on a 4 GB GPU taught me more about LLMs than any course, tutorial, or benchmark ever could.