All Posts
TeamEngineering SpotlightHealthTechCloud Architecture

Engineer Spotlight: Jagdish Parmar

JRHMSquare Team· Team SpotlightOctober 27, 20262 min read

Healthcare software has a problem that doesn't show up in most product demos: the data a doctor needs is often locked inside a system that was never built to let it out. Getting it moving — safely, and without asking anyone to retype it — is the kind of work Jagdish Parmar does.

Seven years from APIs to architecture

Jagdish started out writing clean-architecture REST APIs in .NET Core, then spent six years at CloudPeak leading backend delivery for healthcare and real-time clients. Today he's a Senior Backend Specialist and Cloud Architect at InstaMD, where he owns systems end to end: scoping and schema design, then the APIs, then the AWS infrastructure they run on.

Getting data out of closed EHRs

Much of his current work is EHR interoperability. Platforms like Practice Fusion and eClinicalWorks hold patient charts, labs, and rosters, but offer no practical export path — so clinics fall back on manual data entry. Jagdish builds enterprise browser extensions that capture that data from the clinician's own session, feeding NestJS services that validate, sanitize, and transform it before it lands anywhere. The whole pipeline runs on HIPAA-conscious AWS infrastructure.

Real-time, because healthcare is real-time

Clinical work doesn't wait. That has pushed him through most of the real-time toolkit: RingCentral for fax, voice, and messaging inside clinical workflows; Twilio Voice and SMS for patient contact in remote monitoring; Agora video for consultations that have to stay stable on weak connections; and WebSocket pipelines carrying live translation so a patient and doctor who don't share a language can still talk.

The infrastructure underneath

The less visible half of his work is keeping all of it up. He has migrated a veterinary telehealth platform off bare servers into containerized AWS ECS clusters, built GitHub Actions pipelines that deploy with zero downtime, and tuned PostgreSQL query plans until timeouts under concurrent clinician updates simply stopped happening.

The takeaway

Integration work rarely looks impressive from the outside. Nobody celebrates a chart that synced correctly. But it's the difference between software a clinic tolerates and software it relies on — and it takes an engineer who is equally comfortable in an API contract, a browser extension, and an ECS task definition.

Building something?

We are a team of full-stack, mobile, and AI engineers who ship production products. Let's talk about your project.

Get in touch