I’m a data scientist.
Biochemist by training, machine-learning practitioner by trade. I build clinical and biological ML pipelines, ship indie products solo, and write open-source tools for data analysts.
What I do
Pipelines for brain-tumor classification from DNA methylation, depression-severity prediction from EHR data, and other clinically grounded ML work.
Solo-built and operated learning platforms: BitSize Learning for code, BitSize Language for human languages. Designed, engineered, deployed, maintained.
Co-authored work in genomics + survival modeling. Published technical reports, manuscripts, and reproducible open-source pipelines.
A multi-brand learning platform serving daily Japanese and French lessons (BitSize Language) and coding/data-science modules (BitSize Learning). Independently designed, built, and operated end-to-end. Recent additions include a Speak Challenge module with browser-based speech recognition and a JLPT-level Japanese grammar drilling app spanning N5–N1.
Django · PostgreSQL · Cloudflare R2 · Stripe · SendGrid · Edge TTS · Google OAuth
End-to-end ML pipeline classifying brain-tumor subtypes and predicting survival risk from CpG methylation profiles
Read case study →Benchmarking and predicting patient-reported depression severity at scale
Read case study →Benchmarking transfer-learning architectures on chest X-ray classification
Read case study →Health-data ML, applied research, or builder roles. Based in West Kelowna, BC.