Nabilah Fitriati@nabilahanftr

S.A.G.E. is an AI-powered governance and academic audit platform that automates curriculum compliance, learning outcome mapping, and teaching material verification for higher education institutions, powered by AMD Fireworks AI server-side inference and the Brave Search API for agentic live-web auditing. # Key Features 1. Role-Based Portals: Custom dashboards for Academic Lead, Coordinators, and Lecturers to submit, monitor, and view audit reports. 2. Autonomous Agent Loop: Uses XML-based tool execution blocks to autonomously request baseline definitions, execute live-web searches, and write final reports to the database. 3. Exclusive Fireworks AI Integration: Rebuilt from the ground up to route all inference through Fireworks AI. 4. Live Web Auditing: Integrated with Brave Search API so the agent can query the internet for the currency and recency of software, libraries, and frameworks mentioned in teaching materials. 5. Server-Side Security: Handles all API requests on the backend to safeguard API keys. # Technology Stack 1. Core Backend: PHP 8.x + MySQL 2. AI Infrastructure: - Inference: AMD Fireworks AI API (accounts/fireworks/models/glm-5p2) - Search Gateway: Brave Search API (Web Audit) # Why PHP & MySQL? (Infrastructure Creativity) Initially, S.A.G.E. was conceived as a no-budget project. We did not have dedicated cloud resources to deploy resource-intensive Python/Flask/Django/FastAPI containers or Node.js microservices. To make this solution viable, we had to be highly creative with our infrastructure. We decided to piggyback on the institution's existing shared website server, which only supported a PHP and MySQL stack. Operating within these strict constraints forced us to build S.A.G.E. entirely in native PHP, showing that robust AI agent systems, server-side inference routing (via AMD Fireworks AI), and real-time live-web auditing (via Brave Search API) can be built efficiently without expensive infrastructure overhead.

To participate in the hackathon, simply click Sign up with AMD. If you are not already a member of the AMD AI Developer Program (ADP), you will need to create an account. ADP members can access AI tutorials, experts, and community support. In addition, new members can claim a free month of DeepLearning.ai pro, $100 in AMD GPU cloud credits, and $50 in Fireworks AI API credits. All hackathon participants will receive on hackathon launch additional compute and API credits for their projects. Ready to build what's next on AMD? Sign up with AMD below to secure your spot.