Normalized claim
Decision-making speed: 90% increase
accelerating decision-making speed by over 90%
Golden Energy Mines (GEMS) scales multi-agent intelligence with Gemini to accelerate executive decision-making
Use case typeExecutive analyticsUpdated 5 days ago
Golden Energy Mines (GEMS) in Indonesia built GEMVIS, a hierarchical multi-agent intelligence system on Google Cloud to unify insights across more than 50 application portfolios and improve executive decision-making. The system uses a Dispatcher Agent with Gemini to route requests to specialized agents, combines on-premises private GPU processing for sensitive retrieval-augmented generation with calls to Gemini Enterprise Agent Platform for higher-level reasoning, and embeds the experience into existing internal solutions.
- Organization
- Golden Energy Mines (GEMS)
- Industry
- Energy & Utilities
- Location
- Indonesia
- Published
- July 2026
Reported outcomes
4,000 users
users servedAdoption & scale
+90%decision-making speed−95.8%data retrieval time
Strategic outcomes
Better decisions & insightCompressed complex insight generation to seconds or minutesScale & capacityEnabled a real-time helicopter view across operationsCost efficiencyShifted analysts from manual reporting to baseline training and data quality work
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Data retrieval time: 95.8% decrease
Reduced multi-operational data retrieval time from two days to under one hour
Normalized claim
Users served: 4,000 users increase
GEMS has effectively lowered the barrier to data accessibility for its 4,000+ users
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- Golden Energy Mines (GEMS)
- Provider
- GCP
- Maturity
- Production
- Linked source
- Google Cloud Customer Stories
Multi-operational data retrieval time dropped from two days to under one hour
Customer identity supportedSource describes one deploymentMaturity supported
Primary read
Use case focus
Showing 3 of 3
- 1Executive analytics
- 2Decision support
- 3Agent orchestration
- Fragmented, siloed insights across more than 50 application portfolios created a cognitive bottleneck for leadership.
- Data retrieval for complex insights could take up to two days, slowing cross-domain decision-making.
- GEMS onboarded Google Cloud infrastructure including Google Compute Engine, Google Kubernetes Engine, Google Cloud Armor, Cloud Monitoring, and Google Workspace integration.
- It built GEMVIS as a hierarchical multi-agent system where a Dispatcher Agent with Gemini interprets intent and routes to specialist agents.
- Sensitive mining data is processed on private on-premise GPU servers for RAG, while Gemini Enterprise Agent Platform is used for higher-level reasoning.
- Executive decision-making speed accelerated by over 90%.
- Multi-operational data retrieval time dropped from two days to under one hour.
- The system now serves 4,000+ users through omnichannel embedding across internal solutions.
Architecture
GEMVIS is a hierarchical multi-agent system led by a Dispatcher Agent with Gemini. It routes queries to specialized agents, uses private on-premise GPU servers for sensitive RAG, and calls Gemini Enterprise Agent Platform for higher-level reasoning. The solution runs on Google Compute Engine and Google Kubernetes Engine with Google Cloud Armor, Cloud Monitoring, and Google Workspace integration.
Sources & evidence1
Evidence: Medium65/100Evidence strength
- Customer explicitly identified
- Deployment status explicitly supported
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: Jul 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
Market context
AI patterns
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.