← case studies
Published
June 2026

How a 50-Clinic Network Cut Appointment Waits 70%

A 50-clinic provider ran phone-only triage with 3-week waits, pushing patients to competitors. A HIPAA telehealth platform with AI triage serves 100k+ patients — cutting waits 70%.

100k+

Onboarded onto virtual care

4–6 months

Implementation Time

$500K+

Project Cost
the challenge

A provider running 50 clinics and 200 physicians was stuck on phone-only triage, with three-week appointment waits sending patients to urgent care and competitors. Any solution had to clear HIPAA, SOC 2 and state privacy rules and integrate cleanly with Epic and Cerner — without adding clinical-workflow friction.

what they built

A HIPAA-compliant virtual care platform: (1) a custom WebRTC video engine with adaptive bitrate, screen sharing and recording, tuned for low bandwidth; (2) NLP-driven AI triage that routes patients through pre-consultation intake to the right specialist with context; (3) HL7 FHIR integration with Epic and Cerner for two-way patient, scheduling and clinical-note flow; (4) HIPAA-compliant e-prescriptions with interaction checks, pharmacy routing and refill management.

Bolt built telehealth to disappear into the clinical workflow. A custom WebRTC engine handled video consultations with adaptive bitrate, screen sharing and recording, engineered for the low-bandwidth connections real patients have. An NLP-driven triage layer ran patients through structured pre-consultation intake and routed them to the right specialist with the right context already attached. The hard part was integration: HL7 FHIR connections to Epic and Cerner moved patient data, scheduling and clinical notes in both directions, so physicians never left their system of record. HIPAA-compliant e-prescriptions closed the loop with interaction checks, pharmacy routing and automated refills. Clearing HIPAA, SOC 2 and state-level privacy was a baseline requirement, not a feature. The platform scaled to 100,000+ patients in 18 weeks, with 85% of consultations going virtual and appointment waits cut 70%.

best fit for

Multi-site healthcare providers stuck on phone-based triage who need HIPAA/SOC 2-compliant virtual care that integrates with Epic or Cerner without disrupting physician workflow.

Ai ROLE
AI runs the triage layer. An NLP model takes each patient through structured pre-consultation intake, then routes them to the right specialist with the relevant clinical context already attached — replacing the phone-based sorting that drove three-week waits. It augments physician decisions rather than making them; the video, FHIR integration and e-prescription layers are conventional engineering.
impact

100k+ Patients Served

Active users on the new platform.

70% Shorter Waits

Appointment waits from weeks to days.

85% Visits Virtual

Share of total consultations; 4.8/5 patient rating.
implementation complexity

Custom WebRTC platform, NLP AI triage, HL7 FHIR integration with Epic and Cerner, HIPAA/SOC 2/state compliance, e-prescription. 18-week build, principal-led.

Dean Sloves

Founder, Bolt Group
Bolt Group
Builds production-grade AI and platform systems for PE, mid-market operators, and consulting firms — agentic AI, predictive analytics, embedded engineering.
Get an intro
Talk to this team
industry
Healthcare & Life Sciences
business organization
Operations
Customer Service
AI TYpe
Conversational AI (Chatbot / Agent)
AI-Accelerated Custom Software
value type
Customer Experience
Time Savings
frequently asked questions
How did a regional healthcare provider use AI triage to cut appointment waits by 70%?

A mid-to-large healthcare provider built a virtual care platform with a custom WebRTC video engine tuned for low bandwidth, plus an NLP-driven AI triage layer that ran patients through pre-consultation intake and routed them to the right specialist with context attached. HL7 FHIR connections to Epic and Cerner moved data both ways so physicians never left their system of record. The platform scaled to 100,000+ patients in 18 weeks and cut appointment waits by 70%.

What AI tools and models powered the healthcare provider's telehealth platform?

The triage layer used GPT-4o via Azure OpenAI for NLP-driven intake. The wider stack included WebRTC, Next.js, React Native, Node.js, PostgreSQL, HL7 FHIR, Twilio, and AWS, combining conversational AI with AI-accelerated custom software.

What results did the healthcare provider achieve?

Appointment waits dropped 70%, 100,000+ patients were served, and 85% of consultations went virtual with a 4.8/5 patient rating.

How long did the telehealth platform take to build?

The platform scaled to 100,000+ patients in 18 weeks, within an overall 4–6 month engagement range.

Who is this AI telehealth approach best for?

Multi-site healthcare providers stuck on phone-based triage that need HIPAA/SOC 2-compliant virtual care integrating with Epic or Cerner without disrupting physician workflow.

Have a similar challenge?

Ask whether this would work for you, or describe what you're trying to solve.
TELL US WHAT YOU'RE EXPLORING