The [email protected] email address still works, probably. But the company behind it has moved on.
When Mindy pulled in $6 million from Sequoia Capital back in February 13, 2024, the concept felt almost quaint in its simplicity: an AI Chief of Staff that lived entirely in your inbox. No app download. No browser extension. Just fire off a message to that memorable email address and watch as an AI assistant handled your research, summarized meetings, or tracked down shopping recommendations. Sequoia partner Roelof Botha led the round, with Founders Fund coming in alongside him, backing a founding team with serious pedigree—Yu Pan, a founding engineer at PayPal and the very first employee YouTube hired.
Fast-forward through nearly two years of relative silence, and Sequoia's own portfolio page tells a different story. Mindy is now described as an "AI-powered creator/brand platform" for influencer marketing. The company's LinkedIn feed talks up "Mindy AI Studio" as "your content creation teammate," dropping references to creative workflows and integrations that sound nothing like the inbox-dwelling assistant investors originally bankrolled.
The old TechCrunch writeup? Still there on Sequoia's page, a kind of digital fossil preserving what Mindy used to be. Or what it said it would be.
Which raises the question that shadows nearly every AI product announced in the past few years: At what point does iteration tip over into wholesale reinvention?
The First Chapter
Early 2024. Mindy's founding trio—Pan alongside Cuong Do and Benoit Berthoux—pitched investors on deliberately staying simple. The email interface wasn't a limitation; it was the entire point. Knowledge workers already lived in their inboxes. Why ask them to learn another tool?
According to TechCrunch's reporting at the time, early users ranged widely: financial analysts, students, even defense contractors. All of them treating email as an asynchronous command line for AI queries.
Pricing was still unsettled—maybe subscription, maybe per-query caps, possibly transaction fees on certain tasks. The team was lean, around 10 people. The pitch made sense: a Chief of Staff for the inbox era, handling the ambient busywork that fragments a knowledge worker's attention.
It was, in its way, a credible gamble. Email remains the closest thing to a universal interface in professional life. AI agents were multiplying fast, but most demanded context-switching—new tabs, new apps, new muscle memory. Mindy's framing suggested meeting users in the one place they couldn't avoid.
What Changed (Without Much Announcement)
Somewhere in the stretch between February 2024 and now, the story rewrote itself. No dramatic pivot announcement. No press release declaring a new direction. Just a gradual shift in how Mindy presented itself to the world.
The company's LinkedIn presence started emphasizing "Mindy AI Studio" and content creation capabilities. References surfaced to something called a "Nano Banana integration"—the kind of product-specific jargon that signals a very different user base. Sequoia's current description—"AI-powered creator/brand platform"—reads like it belongs to an entirely separate company.
The email assistant may still function. That [email protected] address presumably hasn't been retired. But the public positioning has migrated decisively toward creators and influencer marketing, a space already thick with tools promising to help content producers scale their output without proportionally scaling their headcount.
There's no roadmap published explaining what happened in those intervening months. The company hasn't issued a mea culpa or triumphant pivot narrative. But the evidence points to a team that met market reality—user behavior that defied expectations, competitive dynamics that shifted faster than anticipated—and quietly adjusted course.
Decoding the Silence
Maybe the email interface, elegant on paper, proved confining in practice. Asynchronous communication has inherent friction. Users expect immediacy now. They want iteration, real-time back-and-forth, the ability to course-correct mid-task. Perhaps early adopters wanted more than glorified research and calendar management.
Or perhaps—and this feels equally plausible—the founding team simply read the room. Grammarly's July 2025 acquisition of Superhuman consolidated a chunk of the email productivity landscape. The AI assistant category became crowded to the point of noise. Creator tools and influencer platforms, meanwhile, were pulling in capital and user attention at a pace that must have looked awfully tempting.
Pivoting toward content creation might have felt less like abandoning ship and more like finally finding the current.
There's also the awkward matter of naming. In November 2025, a different company—1mind—raised $30 million for a sales agent also called "Mindy." Not ideal when brand clarity can make or break product adoption in overstuffed markets.
The Bigger Picture
Mindy's quiet transformation tracks with broader patterns in how AI products are actually developing, as opposed to how pitch decks say they'll develop.
The companies gaining traction lately aren't the ones that clung white-knuckled to their seed-stage vision. They're the ones iterating faster than they're fundraising, chasing user behavior into territory that makes slide decks messy but products useful.
Sequoia's decision to update its portfolio language—while keeping that old TechCrunch link as a kind of historical marker—suggests the firm is comfortable with this fluidity. Worth noting: Sequoia backed Mindy through its Arc program in 2023, betting on the team's ability to navigate fog, not on any single product hypothesis. That bet may yet pay off, even if the product resembles nothing from the original deck.
For founders working in this space, Mindy offers something useful: a case study in adaptive strategy when adaptation isn't optional. The email-first AI Chief of Staff was defensible in February 2024. By now it may have proven to be the wrong idea, executed by the right team.
In venture-backed startups, that distinction matters more than founders typically want to admit. Sometimes the win isn't building what you said you'd build. It's building what users actually need, even when that means leaving your original pitch behind.
