MCP adoption accelerates among enterprise engineers
Dex (@dexhorthy) observes that enterprise engineers shifted from having no MCP strategy in Jan/Feb to routinely driving coding agents like Claude/Codex to use Gdocs, Figma, Glean, and Atlassian MCPs in every workflow within three months.
Score breakdown
If you lead dev productivity at a large org, treat MCP integration with tools like Figma, Gdocs, Glean, and Atlassian as a near-term priority — engineers are already expecting it, and teams without it risk falling behind peers who are getting richer agent context in every coding workflow.
- 01Observer: @dexhorthy, watching engineers at public (large) companies use AI coding tools over the past three months.
- 02In January/February, no engineers had a clear MCP strategy in place.
- 03By April, most engineers are expecting and driving coding agents to use MCPs as part of every workflow.
Twitter user @dexhorthy shares a firsthand observation about a significant behavioral shift among engineers at large public companies over the past three months. In January and February, he notes, essentially no one had their MCP (Model Context Protocol) story figured out. By April, the norm has flipped — most engineers now expect and actively drive coding agents like Claude and Codex to pull in context from enterprise tools including Gdocs, Figma, Glean, and Atlassian via MCP connections as a standard part of their development workflow.
Dex characterizes the adoption curve as a pleasant surprise: slower than he personally would have preferred, but faster than he actually anticipated.
Dex characterizes the adoption curve as a pleasant surprise: slower than he personally would have preferred, but faster than he actually anticipated. The practical implication is that MCP integration with workplace tooling is rapidly becoming a baseline expectation on engineering teams at large organizations, not an experimental edge case. He closes with a direct call to action aimed at dev productivity leaders at big companies who haven't yet addressed this, urging them to get it sorted.
Key facts
- 01Observer: @dexhorthy, watching engineers at public (large) companies use AI coding tools over the past three months.
- 02In January/February, no engineers had a clear MCP strategy in place.
- 03By April, most engineers are expecting and driving coding agents to use MCPs as part of every workflow.
- 04Tools being integrated via MCP include Gdocs, Figma, Glean, and Atlassian.
- 05Coding agents referenced are Claude and Codex.
- 06Dex describes adoption pace as 'slower than I would have wanted but also faster than I expected.'
- 07Dex recommends dev productivity leaders at large companies prioritize establishing their MCP strategy.