What challenges did Promptly face before AI implementation?

Specialty medical practices relied on fragmented tools for scheduling, billing, patient messaging, insurance eligibility, and prior authorizations. Promptly's support and sales teams had to search manually across scattered documentation to answer questions. Developers lacked a shared AI assistant and consistent security posture for software development. The business needed to maintain security guardrails for protected health information while scaling AI experiences.

What was the impact of the AI solution?

The Amazon Bedrock chat is live in production. An Amazon Bedrock AgentCore-powered version is entering beta with multiple clinic customers.

Source

ProductionEvidence: Medium65/100

Promptly (with DoiT): AI-first healthcare patient engagement using Amazon Bedrock and Amazon Q Developer

Use case typePatient engagementUpdated Jun 13, 2026

Promptly is a healthcare software-as-a-service company that helps specialty medical practices streamline patient engagement and automate clinical workflows. The company consolidated tens of thousands of documents from about 50 shared drives into an environment for generative AI applications and agents. Promptly integrated Amazon Bedrock into its web application to provide a conversational chat experience for patients. The chat helps patients with appointment scheduling, billing questions, and insurance questions without requiring a staff member. Promptly also uses Amazon Q Developer across its engineering sprints for code review and software development assistance. An internal assistant called the Oracle of Promptly was built for staff using AWS agent capabilities.

Organization: Promptly
Industry: Healthcare
Location: United States
Published: April 2026

Reported outcomes

Time: Approximately 25% lower

Time & speed

Catalog median for time & speed deployments: −50% across 295 reported metrics. Compare benchmarks →

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Quantified impact: 30-40%

AWS Customer StoriesApr 29, 2026Customer storyInferred claimMedium evidence strength

Amazon Q Developer accounts for 30% to 40% of code written across the engineering team.

Normalized claim

Time: 25% decrease

AWS Customer StoriesApr 29, 2026Customer storyInferred claimMedium evidence strength

Promptly reports about a 25% improvement in development speed and faster movement to beta.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Promptly
Provider: AWS
Maturity: Production
Linked source: AWS Customer Stories

The Amazon Bedrock chat is live in production

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 5

1Patient engagement
2Workflow automation
3Developer productivity

Specialty medical practices relied on fragmented tools for scheduling, billing, patient messaging, insurance eligibility, and prior authorizations.
Promptly's support and sales teams had to search manually across scattered documentation to answer questions.
Developers lacked a shared AI assistant and consistent security posture for software development.
The business needed to maintain security guardrails for protected health information while scaling AI experiences.

DoiT helped Promptly consolidate internal documentation into a centralized environment for generative AI applications and agents.
Promptly built a customer-facing conversational interface with Amazon Bedrock and embedded it in the web application.
The AI chat guides patients through scheduling, bill inquiries, and insurance questions.
Promptly adopted Amazon Q Developer for engineering productivity, using it throughout each sprint and for immediate code review before QA.
The team also built an internal assistant named the Oracle of Promptly using AWS agent capabilities for team-wide knowledge access.

Technologies

Amazon Bedrock Amazon Q Developer Amazon Bedrock AgentCore Amazon Bedrock Agents

The Amazon Bedrock chat is live in production.
An Amazon Bedrock AgentCore-powered version is entering beta with multiple clinic customers.

Architecture

DoiT connected tens of thousands of documents from approximately 50 shared drives into an environment for generative AI applications and agents. Promptly deployed a customer-facing chat widget in its web app using Amazon Bedrock. Engineering teams use Amazon Q Developer in sprint workflows, and an internal assistant was created using AWS agent capabilities. A version using Amazon Bedrock AgentCore is in beta.

Sources & evidence1

Evidence: Medium65/100Evidence strength