I use data to open futures for the people most systems overlook.

A transformer language model, built from scratch

Eight weeks as an Open Avenues Build Fellow, building and training a GPT-style model end to end, from the tokenizer and byte-pair encoding through the training loop and deployment. I wanted to know how the thing works at the level of the code rather than the level of the diagram.

Family engagement and equity

Generalized linear models across 25,391 students, asking whether family engagement protects students differently depending on income.

It does. The protective effect is 18 percent stronger for low-income students, at p = .009, and it held across four separate model specifications, which is the part that made me believe it.

Read the case Code on GitHub

Poverty and gun violence

Census data merged with more than 100,000 incidents across all fifty states, tested with permutation tests, a 10,000-sample bootstrap, and Welch's t-test.

Poverty predicts the rate of incidents. Population does not, which is the assumption most people start with.

Read the case Code on GitHub

Alzheimer's pattern discovery

A knowledge-discovery pipeline over 2,149 patients, reaching 93.8 percent accuracy with an interpretable decision tree rather than a model nobody can read, and validated three different ways.

In clinical work the interpretability is not a nice extra. A doctor has to be able to see why.

Read the case Code on GitHub

Weather ETL pipeline

A five-phase pipeline moving more than 96,000 records into SQLite and BigQuery, with a live Looker dashboard on the end of it. Ordinary engineering, built properly, which is most of what the job really is.

Read the case Code on GitHub

Nothing Crashed

A talk about silent integrity failures in a public data pipeline, where a reference table can be wrong for months and never raise a single error. It covers the threat model, what detection coverage looks like in practice, and what it costs to put in.

I submitted it to the Women in Cybersecurity conference and I am looking forward to hearing back.

National curriculum research

At the Kenya Institute of Curriculum Development, through the country's move from the 8-4-4 system to CBC, including work on special needs education. It is where I learned what happens when a decision made in a spreadsheet reaches every school in a country.