Confidential ai platform
Decision-intelligence SaaS with a two-layer multi-tenant security model
What needed solving
A previous team's Supabase Edge Function build could not handle the complexity of a proprietary decision-scoring framework or the security a multi-tenant enterprise product demands.
How I built it
Rebuilt from scratch: a NestJS backend enforcing tenant isolation at two independent layers (middleware + Supabase RLS), a composite decision-quality scoring engine, and 10 stateful branching simulations with save/resume.
The details
Enterprise decision-intelligence SaaS built around a proprietary 5P framework. Multi-tenant with org_id enforced at both NestJS middleware and Supabase RLS layers, a composite DQI scoring engine, branching simulations, and live executive dashboards.
- Two-layer tenant isolation: NestJS middleware + Supabase Row-Level Security
- 5P Decision Engine with composite DQI auto-scoring
- 10 branching simulation scenarios with stateful save/resume
- Live executive dashboard over Supabase Realtime
- Stripe subscription billing with webhook handling
We can build something that lasts.
Tell me what you are building. I will come back with an honest take on approach, scope and timeline, usually within 24 hours.