原文 · 未翻译
Salesforce has been building software for more than 25 years. It maintains a codebase that stretches across decades featuring every kind of system a modern tech company can accumulate. Thousands of engineers across the globe ship new products on top of it every day.
That's what makes the rapid adoption of Cursor inside Salesforce so notable. When an organization this central to the industry shifts how it builds software, it points at where the whole field is heading.
Today, thousands of engineers at Salesforce use Cursor as part of their daily workflow. Salesforce has already seen double-digit gains in velocity and code quality.
I would say that it's 0 to 1 in terms of how Cursor has transformed the way our developers use tools to improve quality of the product.
#Two paths toward Cursor adoption
Before Cursor arrived, Salesforce had already invested in its own internal AI tools. But Salesforce wanted its engineers to have a range of options, so it made Cursor available.
Junior engineers were the first adopters. Many had started their careers during the pandemic, when remote work made standard ways of learning a codebase unavailable. Cursor helped them catch up.
They didn't have senior engineers sitting with them and explaining a lot of things. Cursor helps them actually understand existing code better so that they are able to contribute more effectively.
For senior engineers, Appajodu observed that they first proved out Cursor on the boring, tedious tasks that are inefficient to tackle manually. And as they experienced the value of automating these tasks and built trust in the tool, these senior engineers expanded their use cases quickly to higher value tasks.
Senior engineers start with the places that feel boring, and once they see the value there, they're ready to use AI coding for other use cases.
Adoption followed the same pattern across teams: a small group would try Cursor, see the impact, and the rest would follow. Within a few months, Cursor went from a new tool at Salesforce to one that was more widely used in day-to-day work. And over time, AI proliferated into every aspect of the SDLC, not just code writing, according to Appajodu.
#Three metrics that matter most
Salesforce is careful about how it evaluates engineering work. The company measures pace and reliability using three key metrics:
Cycle time
Quality (bug count)
Throughput
Salesforce already had dashboards that tracked these metrics across thousands of engineers, making the impact of Cursor visible almost immediately. Cursor helped Salesforce make large improvements in all three areas. One team at Salesforce also reduced legacy code coverage time by 85%.