Production Cloud Cost · 1 of 2

Platform Cost — the shared floor Production-grade (HA / SLA) fixed infrastructure · present whether you run 1 store or 100 · scales step-wise, not linearly

Production HA / SLA Event spine all data sources Pilot floor ~$3.0K/mo

What the ~$3.0K/mo production floor buys (pilot scale)

Component	$ / mo	What it covers
Compute — AKS / Container Apps	~$600	Hosts the ~12 .NET services; orchestration in-process over the event spine
Azure Data Explorer (ADX)	~$1,200	Production time-series cluster — every event + the LLM / decision audit trail
Event Hubs (Standard)	~$100	The event spine — POS, traffic, inventory, sensors, video metadata
Cosmos DB (provisioned)	~$250	Operational records, agent memory, tenant / store context
Cache for Redis (Standard C2)	~$115	Hot state, cooldowns, the statistical anomaly baselines
Azure AI Search (S1)	~$250	Vector index for RAG — store & enterprise knowledge base
Blob storage	~$80	Canonical audit — full LLM prompts / responses retained (SOC 2)
Platform overhead	~$400	Monitoring, Key Vault, identity, networking, egress, contingency
Production floor	~$3.0K	≈ $36K / yr at pilot

What the floor carries — at near-zero marginal cost

The event spine already ingests POS, inventory, people-counters, sensors, weather, supply-chain & staffing. These structured streams are tiny next to 12M video frames/store·mo — video is ~90% of per-store cost, so the retail data rides essentially free on the same infrastructure.

Scales step-wise, not linearly

$3.0K→$4.0K→$18K1 · 10 · 100 stores

Floor grows by tier bumps (AI Search, ADX, Event Hubs Premium) as you add stores — so the per-store share of the floor keeps falling.

Not in this estimate — budget separately

Non-production envs (dev / test / staging)
Integration engineering (POS / inventory / counters)
Third-party data & API fees
Edge hardware (cameras / NVR)
One-time development & the team
Support & SOC 2 / compliance

Production Cloud Cost · 2 of 2 · Conservative basis

1-Year Run Cost — scaling with stores Azure-hosted · per-store LLM agent · frame-sampled video · budgeted on the conservative (per-frame frontier) basis

Video 1 fps · 8 cams Vision frontier escalation Reasoning tiered Haiku→Opus Detection gates + minutely scan

Total cost of ownership — scaling with store count

	1 storepilot	5 storesexpansion	10 storespilot target	→ 100fleet · curve
Platform floor /mo (slide 1)	$3.0K	$3.3K	$4.0K	$18K
Per-store variable (~$1.7K/store·mo)	$1.7K	$8.4K	$16.7K	$167K
Total / month	~$4.7K	~$11.7K	~$20.7K	~$185K
Budget / year plan on this	≈ $56K	≈ $140K	≈ $250K	≈ $2.2M
↓ Optimized target / year	≈ $44K	≈ $81K	≈ $130K	≈ $1.0M

Budget on the conservative basis — ~$1.7K/store·mo (all-in $4.7K → $1.85K/store as you scale). The recommended 3-stage design (clip batching → Gemini Flash + edge-CV) is the path to the optimized target — roughly half — but we hold it as validated upside, not budget, until a footage spike proves the recall.

Per-store cost stack — conservative (~$1.7K/mo)

Vision 1 — Flash-Lite triage, every frame~$410

Vision 2 — frontier on ~3% flagged frames~$1,090

Agent reasoning + minutely scan~$90

Infra + audit substrate (per store)~$80

RAG embeddings~$5

Optimization headroom — upside, not budgeted

Budget basis (this estimate)~$1,670 /store·mo

Optimized target (validated)~$685 /store·mo

The 3-stage cascade (clip batching → Gemini Flash + edge-CV) is ~half — held as upside until a footage spike validates Flash recall. Cost can also rise: 24/7 cameras ~2×, higher frame rate ~6×.