A small FastAPI workout-tracking API whose real job is to be the application I deploy while studying for the CKAD. Multi-container pods, init containers, probes, ConfigMaps, Secrets, Ingress, Jobs and CronJobs all become features of a working app instead of dry exercises.

A simple RAG (Retrieval-Augmented Generation) implementation for practice to augment information generated by LLMs.

Modernized Fedora's release scheduling tooling through the Outreachy internship program. Migrated the codebase from Flask to FastAPI, integrated OIDC authentication, and containerized for OpenShift. 110 commits over 15 months — selected from 100+ applicants, then invited back as a co-mentor.

ML model for predicting diabetes progression, served via a REST API.