Zoom Phone API → SharePoint transcript pipeline (Python, Microsoft 365)

We need a reliable pipeline that pulls call recording transcripts from the Zoom Phone API each day and delivers them into a restricted SharePoint Online library, organised by team and by ISO week, in a defined file format. What you'll build - Extraction service against the Zoom Phone API using Server-to-Server OAuth — paginated, idempotent, resumable, with a daily reconciliation check that alerts on any gap - Delivery into SharePoint Online (Microsoft Graph or PnP PowerShell), foldered by team and week, with a per-week manifest CSV - A team lookup applied at run time and resolved against the call date, so staff who change teams keep their history in the right place - Scheduling, alerting, Microsoft Purview retention labelling, a one-off historical backfill, and a runbook Must have - Production experience with paginated REST APIs, OAuth token refresh, rate-limit backoff, and resumable batch jobs - Python or C#/.NET - Microsoft 365 / SharePoint Online configuration and the Graph API - Scripted, version-controlled provisioning — not manual portal clicking - Clear written English and a working repository we can look at Not required No machine learning or data science. No front end. The analysis is handled in-house — please don't propose it. Non-negotiables All data stays in our Microsoft 365 tenant. No third-party analytics platforms. No production data on your machine at any point — development is against synthetic test data. Start your reply with the word RECONCILE so we know you've read this. We'll share the full scope of works, including the output specification, with shortlisted applicants.