From fff5afb2a3a19e257fd3d8e5ff2f4c598fde98d1 Mon Sep 17 00:00:00 2001
From: Eratostenes de Gitjabia <eratostenes@blocaolabs.com>
Date: Sat, 9 May 2026 12:33:09 +0000
Subject: [PATCH] docs(products): Blocao product overview

---
 docs/02-products/blocao-overview.md | 63 +++++++++++++++++++++++++++++
 1 file changed, 63 insertions(+)
 create mode 100644 docs/02-products/blocao-overview.md

diff --git a/docs/02-products/blocao-overview.md b/docs/02-products/blocao-overview.md
new file mode 100644
index 0000000..231579b
--- /dev/null
+++ b/docs/02-products/blocao-overview.md
@@ -0,0 +1,63 @@
+# Blocao · product overview
+
+## Pitch
+
+Blocao is a sovereign edge AI platform for video forensics. Cameras stream to on-site compute (the Cell), where Frigate detects events and an enricher produces embeddings. An operator console at `http://blocao-router.local/` lets investigators query the system in natural language and play back evidence.
+
+Configuration is GitOps. Network is segmented (VLANs + firewall). Multi-site deployments use a hub on EU sovereign infrastructure. Nothing on US clouds.
+
+## What you get
+
+- **Operator console** with SYNOPSIS landing view (SCADA-style live diagram), CAMS panel, FRIGATE config, MQTT panel, GITOPS panel, HEALTH panel, FORENSICS query.
+- **First-boot wizard** for installer (region, site identity, WAN, camera VLAN, hub, operator).
+- **Camera onboarding** with auto-discovery (DHCP + ARP + ONVIF + RTSP probe), default-credential detection, security checklist, force-password-rotation.
+- **Frigate integration** with detection events on MQTT.
+- **Evidence playback** with clip + correlated metadata + pin-to-case.
+- **Multi-site fleet management** via the hub.
+
+## Differentiators
+
+| Vs | Blocao differentiator |
+|---|---|
+| **Verkada / Rhombus** | On-site storage + sovereignty + open-source stack |
+| **Milestone XProtect** | Modern AI + Linux-based + GitOps + lower TCO |
+| **Frigate DIY** | Production-grade integration + fleet management + UX |
+| **Hikvision / Dahua** | No supply-chain risk + modern AI + fleet orchestration |
+
+See `../05-comparatives/` for detailed comparisons.
+
+## Target hardware
+
+| Component | Default | Alternative |
+|---|---|---|
+| Router | GL.iNet GL-MT6000 | Banana Pi BPi-R4 |
+| Cell | Banana Pi BPI-W3 (RK3588) | Radxa Rock 5B+ |
+| Core (optional) | Jetson Orin NX 16GB | — |
+| Switch (when needed) | Mikrotik CRS305 | Mikrotik CRS326 |
+| Cameras | ONVIF Profile S any vendor | Hikvision/Dahua/Reolink/Axis tested |
+
+## Pricing model (placeholder)
+
+- Hardware: pass-through cost + integration margin.
+- Console + edge AI stack: free.
+- Hub access: per-site monthly subscription (€20-50 range).
+- Support contract: optional, tiered.
+- Custom AI models: project-based.
+
+## Sales motion
+
+Land via:
+
+1. **Pilot install**: 1 router + 1 Cell + ~4 cameras at customer's site. Free first month.
+2. **30-day eval**: customer uses it on real cameras alongside (or replacing) what they have.
+3. **Conversion**: hardware + first-year subscription + support starter.
+
+Expansion via:
+
+- More cameras/Cells at the same site.
+- Adding sites (multi-site fleet).
+- Adding products (transcript forensics).
+
+## Status
+
+In design. Mockups complete for all key views. Sprint backlog ready. First Sprint 0 starts when team is in place.