OpenAI integration.

Define source document identity, namespace/project boundaries, and model/version context. Decide whether the integration moves metadata or derived data; inference and deployment are separate operations. Confirm the connection and read/write operations with Stacksync. Start with one workflow, agree on which fields each system owns, then test initial data, later changes, and recovery before expanding.

Write-back support is not established for this connection. Review the OpenAI records and fields you need with a Stacksync engineer, then test the required operation with a representative record.

Keep provider/project, model or deployment ID, and version distinct.

Resolve project, access policy, model version, and any evaluation requirements.

Test a retired model, a renamed deployment, and an unauthorized project reference. A changed or deleted source updates its intended derived representation, with the correct version and visibility boundaries.

Estimate the number of Models, Files and Audit logs records, the history to load, ongoing changes, and required environments. Include API access, the integration service, destination storage or processing, and the work to maintain mappings and recover from errors. Review current provider pricing against that workload, including any account or API upgrades.