bradleypauld@gmail.com · github.com/bradleypaul · Harding University — B.A. Computer Science, 2015

Pre-seed restaurant management and ordering platform. Partner with the CEO on frontend architecture, data modeling, and product direction. Designed and operate an AI/LLM pipeline that takes a Jira ticket through implementation, tests, and a PR.

B2B industrial e-commerce. Led React/TypeScript modernization (and a Vue 2→3 companion migration) and A/B tested checkout components to reduce cart abandonment.

Reusable React component library and checkout UI for Amazon retail, aligned to the UX system. High unit and integration coverage on flows used by millions of shoppers.

Primary owner of frontend architecture and product direction. Mentored juniors, cut typical ticket time from 5 days to 3, and migrated five core REST endpoints to GraphQL.

Business-formation SaaS. Complex React/TypeScript and Apollo/GraphQL on the client; Node/Express microservices on GCP with PostgreSQL and Kubernetes.

Large-scale transaction and payment systems. React and Vue clients with Redux, Node payment microservices, and REST APIs (Express, with supporting Python services).

Fullscreen Reddit slideshow in the spirit of redditp. Multi-subreddit routes, image and video (including Reddit-hosted, YouTube, and RedGifs), preload, and an NSFW preference that survives refresh. Vite proxy in dev; Netlify rewrite rules for production.

Python client for the Jira Cloud REST API. When you move an assigned ticket to In Progress, a Linux watcher clones the linked GitHub repo and checks out a branch. Also reads issues, comments, and JQL, and can ask Claude for an implementation plan.

Python utility for Apple Silicon: watch a USB HDMI capture and fire macOS notifications when motion appears in a corner ROI — useful for Teams toasts on a capture feed.

React photo gallery built to show component structure and frontend judgment. Written approach covers caching, upload limits, and responsive image serving as follow-ups.

MUI billing card composed from Card primitives and React context instead of prop drilling. Keyboard-submittable, with a write-up of tradeoffs and where AI helped.