Deep Dive · 2025
Aravind — Triage Broker
HIPAA-compliant patient queue routing at scale
ROLELead Systems Architect
DURATION7 months
TEAM1 designer · 3 backend · 1 devops
TIMELINE2025
NestJSKafkaPostgreSQLAWS VPCDocker
Challenge & Limitations
The operational hurdles, data boundaries, and strict service Level agreements of the system.
The Problem Statement
A network of hospitals suffered from bottlenecked admissions. Triage routing was manual, causing delays in ICU admissions and load imbalances across departments. They needed a high-throughput transaction coordinator that automatically routes patient triage details in real time.
Rigid Constraints
- ·Strict HIPAA compliance and data encryption at rest and in transit
- ·Sub-second routing delivery SLA
- ·Integration with legacy HL7 medical feeds
FIG_01: Distributed Triage Data Flow Topology
Design & Engineering Decisions
The tactical solutions mapping backend systems, developer experience, and interface layout.
Architecture
- →Modeled admissions as immutable event streams inside Kafka, partitioned by hospital department
- →Set up NestJS microservices in Docker containers on AWS ECS
- →Implemented secure TLS endpoints for legacy terminal hookups
UX & Interface
- →Designed a high-visibility queue workspace panel for nurses and emergency technicians
- →Used clear status indicators to draw attention to critical triage level updates
Performance
- →Utilized database index partitioning on PostgreSQL for rapid archive lookups
- →Batched telemetry metrics into Redis to avoid heavy read operations during shift turnovers
Deliverables
The Outcome & Business Impact
The triage broker handled over 250,000 daily telemetry events. Admission routing latency dropped from 14 minutes to 800 milliseconds, balancing ICU intake.
<800msRouting SLA
-72%Bottlenecks resolved
250K+Daily events
Navigation