← All cases
Batch pipelines · retail
Content for Sportmaster and X5
Sportmaster · X5 Retail
- Method
- batch pipelines
- food photography and img2video
Task: produce card and campaign material at catalogue scale — batch food photography pipelines and img2video.
How it works
- 01Prompts15+ iterations with the character and the locations pinned
- 02ControlUnit tests against quality regression
Particulars and difficulties
A batch has to look like one campaign
The character and the set of locations are pinned across the whole batch
Prompt work is engineering
Fifteen-plus iterations, each one kept rather than rediscovered
A model update can quietly spoil the output
Unit tests on quality regression catch it before a batch goes out
Different jobs need different models
Several model families in the pipeline, chosen per job
Results
- 15+
- prompt iterations kept
- 4
- model families in the pipeline
- 2
- retail chains
Catalogue-scale imagery
Card and campaign material without booking a studio for every SKU
Quality does not drift
Regression is caught by tests rather than by the client noticing
Stack
Seedream · Nano Banana 2 · Sora · Kling