CONTENT CREATION

Batch-process the Airtable restoration queue on a schedule

On a nightly schedule, pulls all 'Pending' rows from an Airtable restoration queue, upscales each source image with Replicate, stores the result in R2.

CategoryContent Creation
Enginesim
Difficultyintermediate
Triggerschedule
Steps6
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerNightly schedule
  • ActionQuery Airtable for Pending rows by priorityAirtableAirtable
  • ActionUpscale and restore each image via ReplicateReplicateReplicate
  • LogicCheck: valid output and target upscale met
  • ActionUpload restored image to R2CloudflareCloudflare R2
  • OutputUpdate Airtable row to Restored with link and dimensionsAirtableAirtable

What it does

This workflow treats an Airtable table as the system of record for a restoration backlog. Each night it processes every row marked Pending, runs the source image through Replicate, archives the output, and writes the result and status straight back into Airtable.

When to use it

Use it when your archive team curates and prioritizes work in Airtable and wants a predictable overnight batch rather than per-file automation. Good for shops that review candidates by day and crunch them off-hours.

How it works

  1. 1A nightly schedule starts the run.
  2. 2Query Airtable for all rows where Status equals Pending, ordered by priority.
  3. 3For each row, run the source image URL through the Replicate upscale-and-restore model.
  4. 4A check confirms the model returned a valid output and the upscale factor met the target; failures are flagged rather than published.
  5. 5Upload each restored image to R2.
  6. 6Update the Airtable row with the R2 URL, new dimensions, and Status set to Restored.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect AirtableBases, tables, views, automations.
  2. 2
    Connect ReplicateImage, video, and model inference.
  3. 3
    Connect Cloudflare R2Object storage, S3-compatible.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.