Nicholas
Garcia-Scholtz

A Java 17 command-line tool that computes the provably optimal schedule for a set of dependent tasks across a fixed number of processors, solving the classic NP-hard task-scheduling problem. It uses a parallel depth-first branch-and-bound search, with a configurable concurrency level, an admissible lower bound, and state deduplication, to minimise makespan while honouring precedence constraints and communication delays. It also offers an optional live dashboard for monitoring search progress.

Working within a team as a full-stack developer to create a platform to track the club’s $30K portfolio and display daily NZX/ASX stock data. Implemented membership sign-ups using betterAuth and Google OAuth along with domain verification. Integrated Stripe payments for event registration and competition access for members. Utilising Payload CMS as a content management system for articles, public bulletins and events.

Designed and implemented a multi-scene JavaFX educational application simulating a courtroom, where users critically evaluate whether an AI-generated artwork is transformative or constitutes intellectual property theft. Developed autonomous AI and human character witnesses, leveraging OpenAI’s GPT API to provide contextualized, character-driven chat interactions. Utilized multithreading and concurrency to manage real-time chat, timers, and audio feedback, ensuring responsive and seamless user experience.