All case studies
Healthcare
Healthcare AI Triage Assistant Generative AI & LLMs
A US senior-care network triaged incoming patient calls with a front desk, an Excel sheet, and a printed escalation tree - so urgent cases sometimes waited and routine ones got prioritised.
Healthcare
Industry
8 tools
Stack
Client
HealthFlow
Industry
Healthcare
Service
Generative AI & LLMs
Stack
Llama-2 (fine-tuned), LangChain, FastAPI
01
Challenge
“A US senior-care network triaged incoming patient calls with a front desk, an Excel sheet, and a printed escalation tree - so urgent cases sometimes waited and routine ones got prioritised.”
02
Build
We built an AI triage assistant on a fine-tuned reasoning model wrapped in strict guardrails - it scores urgency against the clinic's own protocol and recommends a next step to the front-desk nurse, never diagnoses, and always defers to a human on borderline cases.
03
Outcome
45% faster average triage, 92% clinician agreement with the AI-recommended tier, and an auditable trail on every assisted call.
Deliverables
What the system does — functionality shipped.
- 45% reduction in average triage time across the network. 92% clinician agreement with the AI-recommended urgency tier. Logged + auditable conversation trail for every assisted triage. Zero AI-only decisions - every call still ends with a human confirmation.
Technologies
Llama-2 (fine-tuned)LangChainFastAPIPostgreSQLRedisHIPAA-compliant hostingAudit loggingReact
Similar Build?
Tell us your workflow. We'll review with this example as context and suggest the best path forward.
More Work
View all
Case study © AorBorC Technologies · aorborc.com · +1 (872) 267-2672 (872-AORBORC)