Normalized claim
Candidate search time: 14 days to 15 minutes decrease
reduced candidate search time from 14 days to approximately 15 minutes
Cosmos Aluminium and LCM Go Cloud: Two Amazon Bedrock agents for HR and accounting document search
Use case typeSearch modernizationUpdated Jun 12, 2026
Cosmos Aluminium, a Greek aluminum extrusion company, worked with AWS Partner LCM Go Cloud to build two specialized generative AI agents on AWS for HR and accounting. The CV Intelligence Application evaluates résumés against job requirements, and the Accounting Manual Application answers questions using SAP ERP manuals. The solution used Amazon Bedrock, Amazon Bedrock Guardrails, Amazon Bedrock Knowledge Bases, Amazon S3, Amazon DynamoDB, AWS Lambda, and AWS Site-to-Site VPN with a bilingual Greek and English interface.
- Organization
- Cosmos Aluminium
- Industry
- Manufacturing
- Location
- Greece
- Published
- June 2026
Reported outcomes
30 minutes to near-instant
manual document search timeTime & speed
14 days to 15 minutescandidate search time+99%accuracy16 weeksproduction timeline25 usersinitial user adoption
Strategic outcomes
Employee experienceReduced need for manual expertise and onboardingSpeed & agilityEnabled near-instant answers for ERP manual lookupsCost efficiencyStreamlined HR and accounting workflows
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Manual document search time: 30 minutes to near-instant decrease
replacing workflows that previously required up to 30 minutes
Normalized claim
Accuracy: 99% increase
with estimates closer to 99 percent in practice
Normalized claim
Production timeline: 16 weeks decrease
The project moved from kickoff to production in 16 weeks.
Normalized claim
Initial user adoption: 25 users increase
quickly adopted by 25 users from day one
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- Cosmos Aluminium
- Provider
- AWS
- Maturity
- Unknown
- Linked source
- AWS Customer Story
No explicit deployment-stage evidence found.
Customer identity supportedSource describes one deploymentMaturity evidence evaluated
Primary read
Use case focus
Showing 2 of 2
- 1Search modernization
- 2AI agents
- HR teams manually reviewed more than 1,000 résumés and needed about 14 days to find suitable candidates.
- Employees searching 170 SAP ERP manuals could wait up to 30 minutes for answers, creating delays and inconsistency.
- Built two generative AI applications on Amazon Bedrock.
- Used Amazon Bedrock Guardrails for safeguards and Amazon Bedrock Knowledge Bases for grounded responses.
- Processed documents asynchronously with AWS Lambda, stored documents in Amazon S3, kept metadata and session history in Amazon DynamoDB, and connected on-premises systems through AWS Site-to-Site VPN.
- Reduced candidate search time from 14 days to approximately 15 minutes.
- Returned near-instant responses from 170 ERP manuals instead of waiting up to 30 minutes.
- Improved accuracy from a 95 percent baseline to about 99 percent.
- The project moved from kickoff to production in 16 weeks.
- Users required little to no training and 25 users adopted the solution from day one.
Architecture
The solution consists of two specialized generative AI applications on Amazon Bedrock: a CV Intelligence Application for résumé screening and an Accounting Manual Application for querying SAP ERP manuals. It uses Amazon Bedrock Guardrails and Amazon Bedrock Knowledge Bases, with document processing handled asynchronously through AWS Lambda, documents stored in Amazon S3, session and metadata stored in Amazon DynamoDB, and secure connectivity to on-premises systems through AWS Site-to-Site VPN. The interface supports Greek and English.
Implementation partners1
Sources & evidence1
Evidence: Medium50/100Evidence strength
- Customer explicitly identified
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: Jun 12, 2026Publisher: AWSEvidence: 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
AI patterns
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.