Appearance
Research spikes
Every research spike behind the architecture decisions, in order. Each grounds its findings in a first-party Microsoft Learn source or a named vendor's own documentation, and each traces forward to one or more ADRs.
| Spike | Topic |
|---|---|
| SPIKE-01 | MAI-Image-2.5 image model |
| SPIKE-02 | MAI-Voice-2 text-to-speech model |
| SPIKE-03 | Tenant and subscription readiness |
| SPIKE-04 | Identity, secrets, security, and responsible AI |
| SPIKE-05 | Cost model and governance |
| SPIKE-06 | Publish-pipeline integration |
| SPIKE-07 | Speech models beyond MAI and native Azure (word-sync and lip-sync) |
| SPIKE-08 | Foundry Local (on-device inferencing) |
| SPIKE-09 | AI Foundry model workloads on Azure Local (Arc-connected, cluster-scale) |
| SPIKE-10 | Latest available in-tenant GPT model |
| SPIKE-11 | A newer Grok than grok-4-1-fast-reasoning |
| SPIKE-12 | Broader image, animation, and video generation alternatives |
| SPIKE-13 | Tenant-wide voice/TTS survey |
| SPIKE-14 | Tenant-wide model and region survey |
| SPIKE-15 | Niche and emerging reviewer models for code and document review |
| SPIKE-16 | Photorealistic virtual-trainer avatar |
| SPIKE-17 | Governing agent MCP tools with an APIM AI gateway |
| SPIKE-18 | Foundry Local on Windows Server, and whether the Arc run-command can install it (track 2) |
| SPIKE-19 | Where the ARM and Kubernetes seam sits for Foundry Local on Azure Local (track 3) |
| SPIKE-20 | Cost-first Azure Monitor, Foundry, Grafana, and hybrid observability |
| SPIKE-21 | Solution observability for Azure AI Foundry |
| SPIKE-22 | The Foundry Local model catalog, and whether tracks 2 and 3 draw from one catalog or two |
| SPIKE-23 | Foundry Local install artifacts, and the real mechanics of Azure Arc run command |
| SPIKE-25 | Hardware sizing and capacity planning for the two local Foundry tracks |
| SPIKE-26 | The cost model for the two local tracks, and where a spend cap can actually be enforced |
| SPIKE-27 | What Azure can and cannot see for the local deployment tracks |
| SPIKE-28 | Networking, ingress, TLS, and storage for Foundry Local on Azure Local |
| SPIKE-29 | Lifecycle, upgrade, and drift for the two Foundry Local tracks |
| SPIKE-31 | Cross-track capability and feature parity across the three deployment targets |
| SPIKE-32 | Every model against every region, and what actually differs between them |
| SPIKE-33 | Which client tools can point at this endpoint, and what each one gets wrong |
| SPIKE-34 | Orchestration options, and which of them can be pointed at this account |