Writing
Essays, notes, and field reports.
Independent research on how machine learning is changing biology — and how the industry around it is changing in turn. Longer essays, short observations, and the occasional argument with myself.
2026
-
The model is not the molecule
A better predictor is not a better drug. The gap between model performance and biological consequence is where most AI-biotech claims quietly break.
-
Evaluation is the product
As model weights commoditize, the durable asset in AI biotech is the loop that tells you whether a prediction was any good.
-
Data moats in AI biotech are wet, not digital
Public datasets are a starting point, not a fortress. The durable advantage in AI biotech is operational: assays, automation, and the discipline to keep provenance.
-
What I mean when I say “AI for biology”
A small taxonomy, because prediction tools, design tools, and lab automation are solving very different problems.