Cost¶
Bottom line
The MVP adds no new infrastructure. It runs on our existing Azure DevOps, self-hosted n8n and Azure OpenAI. The only new spend is Azure OpenAI tokens — ~$30–50/month.
MVP monthly (new spend)¶
| Component | How it's provided | New $/mo | Note |
|---|---|---|---|
| Wiki | Azure DevOps Code wiki (Git) | 0 | Included in ADO |
| Orchestrator | Existing self-hosted n8n | 0 | We add workflows |
| Event delivery | ADO service hooks + n8n | 0 | No bus needed at this scale |
| LLM (drafting) | Existing Azure OpenAI · gpt-4o-mini | 30–50 | Pay-per-token, low volume |
| Secrets | Key Vault (likely existing) | ~0 | Managed identity |
| Total | 30–50 |
Scale-up path¶
| Phase | Adds | $/mo |
|---|---|---|
| Phase 2 RAG | Azure AI Search Basic | +75 |
| Phase 2 RAG (thrifty) | pgvector on existing Postgres | +0 |
| Full enterprise (optional) | dedicated AKS n8n, Service Bus, AI Search S1, private networking | 3,000–4,500 |
The enterprise build is optional
It is not required for the four-project scope. We start lean and only scale components that earn their cost.
Effort & payback¶
- Build: 1–2 weeks to a live demo (much of the code already exists).
- Run: low — reviewing PRs is spread across owners; no dedicated compute to operate.
- Cost controls: token quotas per workflow, monthly budget alerts, circuit breaker, cheap-model default (strong model only for hard cases). Confidential/tombstone paths use zero tokens.