CONTENT CREATION

Process an Airtable queue of product images on a nightly batch

Runs nightly over an Airtable table of pending product images, upscales and cleans each one through Replicate, attaches the finished asset back to the row.

CategoryContent Creation
Enginesim
Difficultyintermediate
Triggerschedule
Steps7
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerNightly schedule
  • ActionQuery Airtable for Pending image rowsAirtableAirtable
  • ActionUpscale each source image with ReplicateReplicateReplicate
  • ActionRemove background with ReplicateReplicateReplicate
  • LogicBranch: valid output to Ready, failures to Needs Review
  • ActionAttach finished asset and update status in AirtableAirtableAirtable
  • OutputPost batch summary to SlackSlack

What it does

Drains a backlog of product images tracked in Airtable, enhancing each one overnight so your team wakes up to a batch of catalog-ready assets attached directly to their records.

When to use it

Use when your image pipeline is record-driven — every product lives in Airtable and you want enhancement to run as a controlled nightly batch rather than per-file, with status tracking and retries baked in.

How it works

  1. 1A nightly schedule kicks off the run.
  2. 2The flow queries Airtable for rows where Status is `Pending` and a source image is attached.
  3. 3For each row, Replicate upscales the source image to the catalog spec.
  4. 4Replicate removes the background and outputs a clean transparent PNG.
  5. 5A branch checks whether the model returned a valid output; failures are marked `Needs Review` instead of Ready.
  6. 6The finished asset is attached back to the Airtable row and Status is set to `Ready`.
  7. 7A Slack message summarizes how many images were processed, failed, or skipped.

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 SlackChannels, DMs, threads, mentions.
  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.