.png&w=256&q=75)
🤝 Top Collaborators
.png&w=640&q=75)
🤓 Latest Submissions
RELAY — Meeting to Document Intelligence
Enterprise teams don't struggle to write proposals. They struggle to coordinate them. Nine people contribute to the average enterprise RFP. It takes 9.3 calendar days to complete. 68% of that time is coordination — chasing colleagues, resolving version conflicts, and rebuilding context for every reviewer who joins late. $725,000 in proposal revenue is abandoned annually by the average organisation — not because the content was bad, but because the process collapsed. (Source: Loopio 2025 RFP Trends Report) RELAY is the coordination layer that existing tools don't provide. Upload any meeting recording — MP4, WAV, M4A, any format. RELAY's backend preprocesses it through FFmpeg (16kHz mono MP3) before sending to Speechmatics Enhanced ASR, which returns a fully diarized transcript with automatic speaker separation. The host then labels each speaker with their name, title, and team — with a "hear 5 seconds" audio preview per speaker so identity is confirmed, not assumed. Once speakers are labelled, the host selects a document type: Client Proposal, Statement of Work, Meeting Minutes, Requirements Document, or Strategy Brief. Total time from audio upload to collaborative document: under 90 seconds RELAY hits four of the five hackathon tracks: Agentic Workflows (autonomous audio-to-document pipeline), Enterprise Utility (documented $570K/year coordination problem for VP Sales Operations), Multimodal Intelligence (audio → structured text → AI generation), and Collaborative Systems (Speechmatics handles audio intelligence, Gemini handles generation, humans handle final judgment).
19 May 2026
SAGE IDE — Adversarial Multi-Agent Code Engine
SAGE IDE is a browser-native AI coding environment built on one radical premise: code quality improves when AI systems argue. Unlike single-agent assistants, SAGE deploys four specialized agents into a live adversarial boardroom. The Architect designs. The Implementer builds. The Synthesizer integrates. The Red-Team attacks — mid-stream, without waiting for a review phase. A challenge bar fires in real time, quotes the vulnerable excerpt, and assigns a severity tag (Critical/High/Medium/Low) with CVE reference. The Nash Crucible then aggregates agent votes, detects equilibrium, and locks the final output as a signed, immutable Artifact. The adversarial layer is the core innovation, not an added feature. Every agent has a permanent seat and the authority to interrupt others mid-stream. The boardroom is the interface — not a sidebar, not a chat panel. Agents generate structured Artifacts (Design Plans, Diffs, Challenge Reports, Locked Outputs) that are reviewable, commentable, and auditable. Built on AMD Instinct MI300X via ROCm: four co-resident Qwen/Llama models sharing 184 GB VRAM through vLLM, locked into dedicated memory regions per agent. Backend: FastAPI + LangGraph (7-node graph) + Celery + Redis for async agent parallelism. Frontend: React + Monaco Editor with live WebSocket streaming and gutter attribution marks per agent. CPU fallback (sage-free) validated for non-GPU environments. Benchmarks: +12.4% HumanEval+, +8.2% SWE-Bench-Lite over single-agent baseline. Better process, not a bigger model. What ships: LangGraph multi-agent engine, WebSocket layer, React IDE with Monaco, Agent Boardroom with live challenge bars, Nash Crucible, and full Artifact system — demonstrated end-to-end, no mocks. GitHub: github.com/realruneett/Sage
10 May 2026
