File encoding pipelines

Transloadit runs uploaded files through encoding steps that resize, compress, or change format. A Beam agent watches for a new upload event, reads the source file details and the destination folder, then triggers the pipeline that a customer approved for that file type. Once processing finishes, the agent records the output location on the linked record and notifies the requesting team. Files that fail encoding, exceed a size ceiling, or carry an unexpected format are held and routed to a person, so no broken asset reaches production without a human deciding what happens next.

Assembly status tracking

Every conversion job in Transloadit reports progress and a final result. A Beam agent polls the assembly a job belongs to, reads its state and any error messages, and applies the rule your team set for each outcome. Completed jobs get logged against the originating request and downstream steps are released; stalled jobs past an agreed wait time are surfaced. When an assembly returns an error the agent cannot interpret, it stops, attaches the raw result, and routes the case to an engineer instead of retrying blindly or marking the work finished.

Import and export steps

Transloadit can fetch files from a source and push finished outputs to storage or a URL. A Beam agent reads the import and export configuration tied to a request, confirms the destination matches what the customer authorized, and moves outputs once encoding succeeds. It writes the delivered file path back onto the record and alerts the owner. If a destination is missing, credentials look invalid, or an export returns a permissions error, the agent pauses and hands the case to an administrator rather than sending files where account permissions do not clearly allow it.