Aziz Ullah@azizkhan22kakar31972

Majlis — Arabic for a council or gathering — is a multi-tenant, multi-agent SaaS platform. Every subscribing organization gets a workspace of specialist AI agents (HR, Finance, Records, IT Support, Sales/CRM, Legal/Compliance) plus unlimited fully custom agents built by talking to a conversational builder — all served by one generic agent engine, not six separate codebases. Each agent is just a row in the database: a system prompt, a model, tiered tool grants, and scoped knowledge collections. Agents ground their answers in the org's own documents through agent-driven RAG — they choose when to search, rather than having chunks force-injected — and if nothing relevant is found, they say so instead of hallucinating. A seventh Main Agent coordinates the team: it delegates single-department questions to the right specialist and synthesizes cross-department requests like onboarding a new hire. Safety is deterministic, not left to the model. Every tool call passes through one choke point with three tiers: reads run autonomously, low-risk writes run and log, and high-risk actions (payments, unreviewed external calls) pause for human approval in an Approvals queue surfaced through Slack. A configurable monetary threshold stops large-value actions automatically. LLM inference runs on Fireworks serverless (gpt-oss-20b for routing and basic agents, gpt-oss-120b for HR and Finance). The RAG embedding model is self-hosted on AMD Developer Cloud via ROCm through an OpenAI-compatible endpoint, producing the 768-dim vectors that pgvector searches in Postgres — so AMD compute sits in the critical path of every grounded answer. Multi-tenancy is enforced at the database layer with row-level security.

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.