Plume IA and Unicorne cut costs 78% with a triage solution on AWS

Led by a multidisciplinary founding team that includes physicians, Plume IA offers an AI-powered clinical-documentation solution that transforms patient conversations into structured notes. Through its work with clinics across Quebec, Plume IA observed a persistent operational challenge: a lack of administrative scalability. Across 30 clinics, teams were managing about 6,000 patient calls every day within a limited 8-hour window. During peak periods, high call volumes contributed to long wait times and slow appointment booking for patients, while placing additional demands on administrative teams.

Expanding call-handling capacity through traditional staffing models would require significant additional investment. And providing the equivalent of one dedicated administrative resource at each clinic would represent about USD $1.8 million in annual staffing costs. Also, adding staff would increase capacity during operating hours, but it wouldn’t address the limitations of an 8-hour service window or fluctuations in call volume. Plume IA envisioned a solution that could help patients schedule appointments more consistently and conveniently while supporting clinics in managing high request volumes.

To bring its vision to life, Plume IA engaged Unicorne, a Quebec-based AWS Advanced Tier Services Partner and the winner of the 2025 AWS Canada Rising Star Consulting Partner of the Year. Using AWS, the team gained access to services with powerful AI agents available off the shelf. Unicorne designed a conversational intake and triage system by using Amazon Connect, which provides agentic AI solutions, and Amazon Lex, an AI chat builder. The system detects patient intent, gathers relevant information, and routes requests appropriately, reducing the cognitive load on staff. It automates nonurgent appointment requests and helps patients schedule appointments even when clinics are closed.

Using AWS Lambda, which provides serverless compute, Plume IA supports demand-based auto scaling. When call volumes surge—for example, after weekends—the system absorbs the increase without performance degradation. When volumes subside, the system scales down, saving costs. Given the sensitivity of healthcare information, security and data protection are paramount. Unicorne incorporated AWS Key Management Service (AWS KMS) to create and control keys and AWS Secrets Manager to centrally manage the lifecycle of secrets. The solution also uses Amazon GuardDuty for intelligent threat detection. Using these services together, Plume IA can protect patient information throughout the interaction lifecycle.

To support repeatable, consistent deployment as the solution expands to additional clinics, the team used AWS CloudFormation to speed up cloud provisioning. Amazon Simple Storage Service (Amazon S3) offers industry-leading scalability, data availability, security, and performance. The system handles messaging, notifications, and decoupled processing by using Amazon Simple Notification Service (Amazon SNS), a fully managed publish-subscribe service for application-to-application and application-to-person messaging. To provide message queuing for microservices, the team relies on Amazon Simple Queue Service (Amazon SQS). By implementing Amazon CloudWatch, Plume IA can monitor and optimize its entire technology stack.

Powered by AI, Plume IA’s triage system helps clinics remove one of the most persistent friction points in primary healthcare access. With all-day clinic availability, patients benefit from shorter wait times and faster appointment handling, especially during peak demand.

Plume IA estimated its AWS operating costs at about USD $400,000 per year. Compared with the approximately USD $1.8 million that is required for the equivalent staffing model, this results in a projected annual savings of about USD $1.4 million. The serverless foundation means that costs remain proportional to actual usage, and the architecture can expand to additional clinics without manual infrastructure provisioning. “The economics of the solution are compelling,” says Marc-Andre Paquin, chief technical officer and cofounder at Plume IA. “We expect a cost reduction of about 78 percent. At the same time, patients are no longer constrained by the traditional 8-hour scheduling window, and clinics can manage thousands of interactions without administrative capacity becoming the bottleneck.” With this foundation in place, Plume IA can continue expanding its healthcare AI capabilities, bringing intelligent automation to additional aspects of clinical administration.

Based in Quebec, Unicorne specializes in migration and modernization, security, managed services, and cost optimization.