i work as a staff data scientist @ 
. my interests:
- nlp
- functional analysis
- probability
- statistics
- walking
- tennis
- chess
- overwatch
- yugioh
- movies
- music
i have a book coming called In ML: preorder ↵
work
staff data scientist @
where i currently work.
- pioneer member of the data team.
- built the data stack from scratch — warehousing, reporting database, etl, and productionization.
- grew the team through hiring, mentoring, okr-setting, and tooling.
- cross-functional work with customer experience, development, and executive teams.
- company-wide financial and customer forecasting (arima, prophet).
- nlp in production — brand/creator matching, similarity scoring, and topic modelling (word2vec, doc2vec, glove, bert, gpt).
- private python libraries and docker containers for model deployment.
- deployment via flask and sagemaker, automated with circleci, helm, and argocd.
- rshiny and flask apps for internal ml servicing.
- custom reforge methods to isolate user behaviour (setup, aha, and habit moments).
- primary owner of a/b test validation and experimental design.
modelling
nlp (word2vec, lda) supervised (regression, trees, neural nets) unsupervised (pca, clustering) monte carlo
programming
r python sql ruby js / react
software
aws (sagemaker, s3, ecr, ec2, lambda) amplitude bigquery gce fivetran segment rshiny flask ga notion zendesk asana lm studio ollama hugging face + more
at the moment, i’m building some new product features.
in ml
Deep Learning with Functional Inputs
deep learning with functional inputs
a framework for feed-forward neural networks that take functional data as input — a scalar response modelled on one or more functional covariates, plus any number of scalar covariates. the learned parameters are themselves functions, which keeps the model interpretable. validated on both predictive accuracy and recovery of the underlying coefficient functions via cross-validation and simulation.
funcnn — an r package
an r package for deep learning with functional and scalar covariates. built on keras/tensorflow, with functions for model building, prediction, and cross-validation, plus documentation of the methodology.
overwatch
grandmaster support
01 / 03
02 / 03
03 / 03
yu gi oh!
some pictures of card arts
yu gi oh!
chess
play ai barinder in chess
i started playing chess a couple of years ago and have been trying to get better; this bot plays kind of like me ;)
