Normalized claim
Time: 6-10 minutes
Certain claims that previously took 6–10 minutes to file can now be processed in under 30 seconds.
stp.one delivers agentic legal AI using Amazon Bedrock (Legal Twin)
Use case typeLegal workflow automationUpdated Jun 13, 2026
stp.one is a Germany-based legal tech software provider serving law firms and notaries with document-heavy case workflows. The company built Legal Twin on AWS to accelerate legal research, litigation discovery, case analysis, document retrieval, invoicing, and collections while meeting GDPR and data-residency requirements.
Reported outcomes
6-10 minutes
timeTime & speed
6-10 secondstime
Strategic outcomes
Speed & agilityAccelerated claims processingBetter decisions & insightImproved document retrieval and legal researchRisk & complianceMet data-residency and legal compliance requirementsInnovation & cultureExperimenting with multi-agent workflows
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Time: 6-10 seconds
Certain claims that previously took 6–10 minutes to file can now be processed in under 30 seconds.
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- stp.one
- Provider
- AWS
- Maturity
- Unknown
- Linked source
- AWS Solutions Case Study
No explicit deployment-stage evidence found.
Customer identity supportedSource describes one deploymentMaturity evidence evaluated
Primary read
Use case focus
Showing 3 of 4
- 1Legal document analysis
- 2Case research automation
- 3Workflow automation
- Law firms handle large volumes of document-driven work across long-running cases, creating bottlenecks in research, discovery, and drafting.
- stp.one needed an AI platform that could comply with GDPR, German lawyer regulations, and customer data-residency requirements.
- stp.one built Legal Twin on AWS and used Amazon Bedrock to access foundation models for agentic AI capabilities.
- The product uses Anthropic’s Claude in Amazon Bedrock for natural-language generation and Amazon Bedrock Titan embeddings for semantic search.
- Legal Twin is integrated with the document management system and uses Amazon Aurora PostgreSQL for high-performance semantic search and vector-store support.
- The platform can also connect on-premises storage with cloud-based processing for customers with sovereignty constraints.
- Certain claims that previously took 6–10 minutes to file can now be processed in under 30 seconds.
- The article says Legal Twin helps users triple productivity at the upper end.
- stp.one reports improved document retrieval, legal research, litigation discovery, and receivables processing.
- The team is using the platform to experiment with multi-agent workflows.
Architecture
Legal Twin is built on AWS around Amazon Bedrock and Amazon Aurora PostgreSQL. It uses Claude in Bedrock for generation, Titan embeddings for semantic search, and Aurora PostgreSQL as the database supporting semantic search and availability. The solution integrates with the customer’s document management system and can connect on-premises storage with cloud-based processing.
Sources & evidence1
Evidence: Medium50/100Evidence strength
- Customer explicitly identified
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Case StudyPublished: Apr 29, 2026Publisher: AWS Solutions Case StudyEvidence: 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.