MED360 Presupuestos: following up every surgical quote, with WhatsApp campaigns and weekly summaries

Scope: operators at Clínica CBB follow up surgical quotes from the admin panel, with WhatsApp campaigns and an AI-generated weekly summary. It is in production.

Sector
Healthcare
Role
CTO at Blanco Brand
My part
Technical direction, as CTO of Blanco Brand.
Period
In production
Status
In production
Team
Ten authors across the repositories

System blueprint

System blueprint: MED360 PresupuestosSurgical quotes enter the system from Excel spreadsheets. The clinic’s operators follow up the quotes. Patient campaigns go out over WhatsApp with Twilio; the API works with queues on BullMQ and Redis. Weekly summaries are generated with OpenAI. Underneath, the platform: a NestJS API, MongoDB and a React admin panel.01Excel spreadsheetsingestion of surgicalquotes02Operatorsfollow-up from the adminpanel03CampaignsWhatsApp with Twilio;queues on BullMQ and Redis04Weekly summarygenerated with OpenAISystem blueprint: MED360 PresupuestosSurgical quotes enter the system from Excel spreadsheets. The clinic’s operators follow up the quotes. Patient campaigns go out over WhatsApp with Twilio; the API works with queues on BullMQ and Redis. Weekly summaries are generated with OpenAI. Underneath, the platform: a NestJS API, MongoDB and a React admin panel.01Excel spreadsheetsingestion of surgical quotes02Operatorsfollow-up from the admin panel03CampaignsWhatsApp with Twilio; queues onBullMQ and Redis04Weekly summarygenerated with OpenAI

The flow, in text

  1. Surgical quotes enter the system from Excel spreadsheets.
  2. The clinic’s operators follow up the quotes.
  3. Patient campaigns go out over WhatsApp with Twilio; the API works with queues on BullMQ and Redis.
  4. Weekly summaries are generated with OpenAI. Underneath, the platform: a NestJS API, MongoDB and a React admin panel.

Platform: NestJS API · MongoDB · React admin panel · Diagram. Not live data.

Product in use

Product interface with demo dataThe operations room: what each operator is working on and which quotes are going unattended.

Problem and constraints

MED360 is a family of patient communication and management products for medical institutions; Presupuestos is one of them. At Clínica CBB, operators follow up patients’ surgical quotes. MED360 Presupuestos is the system where they do that work.

Quotes arrive in Excel spreadsheets, and contact with patients happens over WhatsApp.

Constraints

  • Quotes enter the system from Excel spreadsheets.
  • Patient campaigns go out over WhatsApp, with Twilio.

Decisions and architecture

Quotes are imported from Excel spreadsheets. Operators follow them up from the admin panel.

The system sends WhatsApp campaigns with Twilio and generates weekly summaries with an OpenAI model.

The API is NestJS on MongoDB, with AWS Cognito, queues on BullMQ with Redis, and Socket.IO for real time. The admin panel is React with Vite.

There is also an assistant that queries the quotes database and exports to Excel. It is a separate service, built with Flask, LangChain and OpenAI.

Results

MED360 Presupuestos is in production at Clínica CBB.

It has a sibling product, MED360 Postconsulta: patient follow-up and reminders after an appointment, with analytics. Postconsulta uses NestJS, MongoDB, Socket.IO, Twilio and SendGrid/SES, with a Next.js web app.

  • 2MED360 products: Presupuestos and Postconsulta.
  • 10authors across the repositories.

Limits

  • This case describes the scope of the system; it does not include measured results.
  • It does not publish metrics for the campaigns or the AI-generated summaries.

Do you have a system that cannot be touched?

Tell me the problem

Want the full career history?

View the CV

Let’s talk. If you want to discuss AI, healthcare, fintech or research, propose a collaboration or invite me to give a talk, write to me.

  • Asturias, Spain
  • CET hours
  • Spanish native, English C1

Also on WhatsApp · GitHub