Bluesight builds agentic AI solution for healthcare compliance using Amazon Bedrock AgentCore
Prism Assistant for ControlCheck launched in May 2026 and is already in use by 20 health systems, with a multi-product agentic solution planned for later in 2026.
1 source · cross-referenced
- Bluesight, a healthcare compliance software provider, used AWS engagements and Amazon Bedrock AgentCore to build Prism, a unified agentic AI solution spanning six healthcare compliance products.
- Prism Assistant for ControlCheck launched in May 2026 and is already in use by 20 health systems.
- A more complex multi-product agentic solution is planned for later in 2026.
- The solution leverages Amazon Bedrock’s HIPAA eligibility, AgentCore Runtime for secure hosting, and AgentCore Gateway for API integration.
Bluesight, a healthcare compliance software provider serving thousands of U.S. partners, built Prism, a unified agentic AI solution spanning six healthcare compliance products using Amazon Bedrock AgentCore. The company collaborated with AWS through two engagements to evolve from a single-product AI prototype to a production-grade system designed to address cross-product compliance challenges.
Prism Assistant for ControlCheck launched in May 2026 and is already in use by 20 health systems. A more complex multi-product agentic solution is planned for later in 2026, according to the announcement. The company’s product suite includes KitCheck, ControlCheck, CostCheck, 340BCheck, ShortageCheck, and PrivacyPro.
The compliance challenges driving Bluesight’s agentic AI efforts included drug diversion detection and GPO prohibition use cases. ControlCheck monitors controlled substance transactions, requiring compliance teams to compile reports and correlate signals across dashboards. For GPO prohibitions, hospitals must prove exceptions by cross-referencing purchase records, shortage data, and 340B eligibility across multiple Bluesight products, none of which had a unified view.
Bluesight chose Amazon Bedrock AgentCore for its production-grade agentic AI infrastructure, emphasizing three capabilities. Amazon Bedrock is HIPAA-eligible, operates under a Business Associate Agreement (BAA) with AWS, and does not use customer data to train foundation models. AgentCore Runtime provides secure, serverless hosting with session isolation for concurrent hospital queries. AgentCore Gateway transforms existing product APIs into Model Context Protocol (MCP) compatible tools with built-in authentication and encryption.
The company’s agentic architecture uses a coordinating agent that delegates to specialized data workers. For the GPO prohibition use case, one worker queries CostCheck purchase records, another checks ShortageCheck availability data, and a third validates 340B eligibility. AgentCore Runtime supports this multi-agent pattern with separation of concerns and observability across the execution chain.
Bluesight partnered with AWS through the Experience-Based Acceleration (EBA) program for a three-day intensive sprint in September 2025 to build a conversational drug diversion analyst for ControlCheck. The team of 15—eight Bluesight engineers and seven AWS professionals—built a functioning agent with Strands Agents on Amazon Bedrock, hosted on AgentCore Runtime. They connected over 10 ControlCheck APIs through AgentCore Gateway as MCP servers and implemented observability for performance monitoring and cost attribution.
A key architectural decision separated AI reasoning from the data layer by wrapping existing ControlCheck API endpoints in AWS Lambda functions that returned structured, agent-optimized data. This reduced query latency from 5 minutes to 10 seconds and kept business logic in the application layer. The agent’s role focused on interpreting user questions, orchestrating tool calls, and presenting results clearly.
The EBA engagement delivered an architecture ready for production from day one, deployed in a virtual private cloud (VPC) with full encryption, authentication, observability, and infrastructure-as-code in place. No architectural pivots or accrued technical debt were reported.
- Jul 12, 2026 · NVIDIA — Deep Learning Blog
NVIDIA Nemotron 3 Ultra achieves leading performance with LangChain Deep Agents harness at lower cost
Trust78 - Jul 12, 2026 · AWS — Machine Learning Blog
AWS SageMaker AI adds serverless fine-tuning for NVIDIA Nemotron 3 models
Trust84 - Jul 12, 2026 · GitHub · vllm-project/vllm releases
vLLM v0.25.0 drops PagedAttention, makes Model Runner V2 default, adds multimodal and speculative decoding features
Trust79