A claim wall for biomechanics
A mechanics engine for lifters is only trustworthy if the outputs it cannot support are unreachable by construction, not merely omitted from the current release.
biomechanicssoftware designverificationdigital twins
Writing and commentary
A public research journal
Paper reactions, research notes, hardware lessons, and commentary on what seems important or overhyped in the current research landscape.
No writing entries match that filter yet.
Every number in the notebook has an oracle
A workshop number only counts once something the author did not derive agrees with it; when it does not, the notebook says so instead of hiding it.
teachingnumerical verificationoptimizationjupyter
Why Gradus refuses to let a language model grade anything at runtime
Gradus grades and diagnoses with rules, not a model, so every outcome is reproducible; the price is a heavy authoring contract and an audit for hollow content.
adaptive learningdeterministic gradingproduct designeducation
Preregistering against yourself: six phases of trying to falsify a memory architecture
Locked memos, validity gates, and paired-seed statistics turned a memory-architecture idea into a program where a negative result is as usable as a positive one.
kernel methodspreregistrationsequence modelingresearch methodologyhpc
The camera push was never on the critical path
The capture loop closed with hand-dialed recipes plus a longitudinal log; the Sony live push arrived last, set four of six fields, and was an optimization.
photography toolingfeedback loopscamera sdkproduct designlocal-first
Source separation for zebrafish calcium imaging
The strongest result was not a single winning denoiser: preserving the activity carrier and using spatial separation features for ranking proved more reliable.
calcium imagingsource separationsignal processingneuroengineering
Why I care about state-space models for EEG
EEG work gets more informative when temporal structure is part of the model instead of something the pipeline averages away.
eegstate-space modelssignal processing
Building an EEG/EMG helmet from scratch
Biosignal research gets cleaner when hardware, firmware, and modeling are treated as one system instead of three separate concerns.
eegemghardwareneuroengineering
Teaching analysis and ML with Jupyter at UF
Workshop design works better when theory and application are taught as one loop instead of two separate tracks.
teachingmachine learningjupyter