Research notes
Read note Reading the geometry of AS-OCT pretraining
Measured representation changes across five AS-OCT pretraining runs, and the profiling work that cut two-GPU JEPA update time by 14.2%.
Writing
Research engineering, worked through.
The clinical AI systems I build and operate, the experiments I run, and the engineering decisions behind them. From OphtaFlow AI's real-time diagnostic support to corneal geometry and AS-OCT learning, these articles explain the work through concrete examples.
Research notes
Experiments, measured results, and the decisions they inform.
From measurements to models
7 connected articles. Read them in order, or start with the problem that interests you.
- Representation learningRead article
Learning without labels: what MAE and JEPA actually predict
The learning objectives I use for AS-OCT: reconstruct hidden pixels, align views of an image, and preserve variation between sources.
- Experimental designRead article
Five OCT pretraining recipes, and the choices behind them
The five AS-OCT recipes I implemented on 3.43 million images: their inputs, objectives, shared controls, and compute tradeoffs.
- Clinical AI systemsRead article
From device data to real-time diagnostic support
How I built OphtaFlow AI for diagnostic support in production 24/7, powered by CorneaForge, and the next stage of its clinical and research platform.
- Scientific computingRead article
From measured points to Zernike coefficients
How I turn native corneal samples into maps and Zernike descriptors in CorneaForge, with explicit reference surfaces, coordinates and units.
- Performance engineeringRead article
Cache the geometry, then change the values
How I accelerate CorneaForge’s geometry engine with reusable interpolation, batched surface fits, and directly evaluated derivatives.
- Scientific computingRead article
From corneal shape to optical aberrations
How I built the optical calculation in CorneaForge: surface slopes, refraction, ray intersections and optical path differences become Zernike descriptors.
- Scientific computingRead article
The reference surface changes the residual
How I implemented conoid and biconic fitting in CorneaForge, and why choosing a reference determines which parts of corneal shape remain in the residual.
See also my peer-reviewed publications, or subscribe to new articles via RSS.