ML Notes
Reading notes and technical reflections on deep learning, representation learning, and computational pathology ML.
These are research-learning notes, not formal tutorials or claims of expertise. I use this section to write down what I am learning from papers, model architectures, experiments, and implementation work.
Model architecture notes
- transformers
- mixture-of-experts
- attention and KV-cache pressure
- long-context inference
- architecture/hardware co-design
Computational pathology ML
- whole-slide modeling
- multiple-instance learning
- pathology foundation-model features
- scanner/site robustness
Experiment engineering
- reproducible scripts
- ablations
- repeated seeds
- baselines
- claim boundaries
Earlier systems/security work
Earlier technical work included home-lab and defensive-security practice. That remains part of my systems background, but my current portfolio focus is computational pathology research engineering and machine-learning research notes.
Older systems/security backgroundThese notes are personal technical reflections and research-learning notes. They are not clinical guidance, deployment documentation, or peer-reviewed claims.