Normalized claim
Quantified impact: 20% increase
Items listed increased by about 20%
GCPEvidence: Medium50/100
Carousell: using Gemini/Vertex AI + BigQuery to automate listings and run bulk fraud inferences
Use case typeDocument automationUpdated Jun 13, 2026
Carousell Group, a multi-category classifieds and recommerce marketplace, used Google Cloud generative AI and data services to improve seller listing creation, scale fraud analysis, and automate internal workflows. The company built a new "List with AI" experience that lets sellers upload a short video or photos and uses Gemini's multimodal capabilities to generate listing titles, descriptions, category, price, condition, and style variants for review. Carousell also moved fraud model bulk inference to an integrated BigQuery and Vertex AI workflow so it can analyze its full chat dataset and iterate much faster than before. Internally, Carousell created a Gemini-based translation workflow and a secure Slack bot that answers HR questions from policies stored in Google Drive.
- Organization
- Carousell Group
- Industry
- Retail
- Location
- Singapore
- Published
- May 2026
Reported outcomes
+20%
quantified impactOther quantified impact
90%quantified impact−10%cost
Strategic outcomes
Better decisions & insightScaled fraud analysis across full datasetSpeed & agilityAutomated most translation workflowsCustomer experience & trustReduced seller listing friction
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Quantified impact: 90%
Automated about 90% of translation needs
Normalized claim
Cost: 10% decrease
Compute cost reduced by about 10%
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- Carousell Group
- Provider
- GCP
- Maturity
- Unknown
- Linked source
- Google Cloud Customer Stories
No explicit deployment-stage evidence found.
Customer identity supportedSource describes one deploymentMaturity evidence evaluated
Primary read
Use case focus
Showing 3 of 4
- 1Workflow Automation
- 2Fraud Detection
- 3Generative AI
- Reduce friction in the seller listing workflow
- Scale fraud detection across a high-volume chat dataset
- Automate manual translation and HR self-service tasks
- Built a "List with AI" feature using Gemini multimodal input from video and photos to generate listing fields automatically
- Used BigQuery and Vertex AI for bulk fraud inferences across the full dataset
- Built a custom Gemini API workflow to automate most translation needs
- Built a secure Slack bot that uses Gemini to answer HR policy questions from Google Drive
- Median listing time reduced to less than two minutes
- Items listed increased by about 20%
- Fraud model iteration cycle reduced from about three weeks to a couple of days
- Automated about 90% of translation needs
- Compute cluster on-call workload reduced to zero
- Compute cost reduced by about 10%
Architecture
Carousell uses Gemini multimodal capabilities for seller listing generation, BigQuery integrated with Vertex AI for bulk fraud inference, and a custom Gemini API workflow plus Google Drive-backed Slack bot for internal automation.
Sources & evidence1
Evidence: Medium50/100Evidence strength
- Customer explicitly identified
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: May 23, 2026Publisher: Google CloudEvidence: PrimaryConfidence: High
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
Browse the catalog
Was this useful?
Community
Comments
Loading comments...