marketing_ops · saas · workflow
Modal uses evals and inference-time compute scaling to build QArt codes with a ninety-five percent scan rate
QArt codes — QR codes embedded in AI-generated images — reliably failed to scan in real-world use despite impressive demos, making it impossible to consistently produce codes that were both beautiful and scannable.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · User request initiates batch
A user request to the service triggers the generation of eight QArt codes in parallel.
Tools used
Stable DiffusionControlNetQReaderYOLOWeights & Biases' WeaveModal
Outcome
By developing automated evals and applying inference-time compute scaling with eight parallel generations per request, the team achieved a ninety-five percent scan rate SLO while improving aesthetic quality and returning codes to users in under 20s, p95.
What failed first
An initial version of the system produced scannable codes only when using simple texture or style prompts, failing with complex scene descriptions, and falling far short of the original viral demo's promise.
Results
Time savedunder 20s, p95
Volumeninety-five percent
Grounding & classification
Source type: technical build writeup
21 fields verified against source quotes.
computer visioncontent generationquality inspectionfailure mode describedmetric backednamed customerproduction runtime claimedtools describedworkflow describedsoftwareaccuracy improvementcycle time reductiontechnical build writeupmarketing ops