Context-aware SQL generation
Turning plain-English business questions into SQL people can actually trust — by giving a language model the schema, the vocabulary, and a reviewer.
- LLMs
- SQL
- Metadata
- Human review
Projects
A mix of professional success stories, open-source tools, and things I built because I wanted to see if I could. Labels are honest: an experiment is an experiment, and nothing here is a product for sale.
Turning plain-English business questions into SQL people can actually trust — by giving a language model the schema, the vocabulary, and a reviewer.
Cutting repetitive manual checks and inconsistent calls in document validation — with OCR, field-level checks, confidence signals, and a human for the hard cases.
A desktop app for early-stage data modeling: load a CSV, explore distributions and correlations, plot, and train and compare scikit-learn models without writing a notebook.
Hands-on experiments in scaling and tuning gradient boosting classifiers on a public health-indicators dataset, written up as a blog post with code on GitHub.
Python helpers for pulling tables out of PDFs — including tables that continue across page breaks — and writing them back out as clean CSVs.
A small wrapper around the boto3 calls I kept rewriting for data in S3 — list keys under a prefix, or read a CSV straight into a pandas DataFrame.
A Go tool that fully automated a recurring AWS database migration chore — saving me about an hour every time data in the cloud needed updating.
An Android app written in Dart and Flutter that serves up dad jokes from the icanhazdadjoke API — guaranteed to make you roll your eyes.
An experiment in interactive streaming for people who like engineering, math, and technology — and want to participate rather than just watch.
FDM printing at home — rapid prototyping, custom parts, and once, a prototype that helped an acquaintance get a product to a manufacturer.
From IPA enthusiast to IPA maker: notes from my first brew day, mistakes included.