$ loadingLoading
projects
Five systems across machine learning, backend, desktop, and mobile. Each one was built to run somewhere real — offline on an Android phone, on a Linux VPS behind nginx, or on a laptop with no network at all. The case studies cover the decisions that made that possible.
A cross-platform desktop app that turns any LLM into a structured, interactive tutor.
lyceum --model local:qwen --topic 'gradient descent'
session ▸ tutor mode: adaptive
provider ▸ local (no key required)
stage 1 ▸ explanation
stage 2 ▸ worked example
stage 3 ▸ check for understanding …
A transport-agnostic AI agent that autonomously diagnoses faulty computers.
A modular AI platform for predicting disease outbreaks across Tanzania.
A multisignature escrow platform securing mobile money for East African merchants.
An ultrasound classifier for breast cancer screening, built to run offline on-device.
Every project on this page links to its repository. Smaller experiments live there too.