Challenge
Exporting a full Dubber archive is slow. The export API returns CSVs by date range, and audio has to be downloaded one recording at a time under rate limits. Regulated firms still need years of recordings kept intact.
Approach
Retrieval and migration pipelines with OAuth, pagination, retries and webhooks. Recordings are delivered encrypted to the client's own S3, R2 or Azure storage, and no audio is kept on our side.
Stack
- Dubber API
- OAuth
- Webhooks
- S3 / R2 / Azure
Highlights
- Migrated ~1 TB (310k+ recordings) for an FCA-regulated fintech, with no audio kept locally
- Full archive and metadata moved into storage the client controls
- Designed around GDPR, PCI DSS, MiFID II and FCA retention rules