Technical deep-dives on Salesforce, Heroku, MuleSoft, Airflow, and AWS from engineers who build with them every day.
A client wanted to clean up 800+ legacy custom fields before a major data migration. The native dependency graph wasn't enough. Here's the script we built — and what it found.
How we built an eight-application MuleSoft integration layer for a digital health company — connecting a customer portal, Salesforce CRM, and a logistics platform with bidirectional case management and automated kit tracking every 5 minutes.
How we built a bidirectional Salesforce ↔ NetSuite integration using MuleSoft API-led connectivity — Platform Events for real-time orders and customers, watermark-based scheduled sync for invoices and refunds, and SOAP/XML for complex Sales Order operations.
How we automated Salesforce-to-EMR patient data sync for a therapy clinic with no API access — Heroku, Puppeteer, and RabbitMQ running 24/7, zero double entry.
Heroku Connect is easy to misconfigure in ways that only surface at volume. Here are the failure modes we see most often and how to address them.
REST, Platform Events, or RabbitMQ — how to choose the right Salesforce-Heroku integration pattern based on team ownership, cost, and failure modes.
The four core Salesforce integration patterns — REST, SOAP, Platform Events, and Change Data Capture — with a decision framework for choosing the right one.
How to trigger an asynchronous process on Heroku or AWS from Salesforce Apex using RabbitMQ — avoiding callout limits and keeping data transfer secure.
How to run Apache Airflow on Heroku to move data from multiple sources into Salesforce using bronze, silver, and gold layers with Heroku Connect.