Inspiration We were both new to HPC and felt lost—GPU/CPU choices, queues, and costs were overwhelming. We wanted a guided, non‑technical path to get started with confidence.
What it does AURORA lets users upload a project, run a quick calibration, and get clear recommendations on the best compute options based on budget, time, and performance—explained in plain language.
How we built it We built a React/Next.js frontend and a FastAPI backend. The backend runs a calibration sample, captures runtime signals, and produces ranked recommendations. The UI focuses on a simple workflow and readable results.
Challenges we ran into Getting local calibration to run reliably for demos, aligning OAuth and environment setup, and turning technical metrics into beginner‑friendly guidance.
Accomplishments that we're proud of A clean end‑to‑end workflow, an AI‑native interface that feels approachable, and a system that surfaces real tradeoffs instead of guesswork.
What we learned Beginner users need clarity over complexity; explainability is just as important as accuracy. Calibration signals are the key to practical recommendations.
What's next for Aurora More cloud providers, deeper performance profiling, team collaboration features, and a fully guided onboarding path for first‑time HPC users.
Log in or sign up for Devpost to join the conversation.