Xobni
Two engineers — one from MIT, one from Penn State — met at a Washington D.C. internship, recognised that email was a broken mess of buried relationships, and built a product from a dorm room that Bill Gates personally demoed at a Microsoft conference. Then they turned down $20M from that same Microsoft — and eventually sold for $60M+ to Yahoo instead.
Adam Smith · 18 min read
Xobni, YC Founder Story
Company: Xobni (pronounced "ZOB-nee", "inbox" spelled backwards) Founders: Adam Smith (CTO) & Matt Brezina (Co-Founder) YC Batch: Summer 2006 Industry: Email Productivity / Developer Software / Mobile Founded: March 2006, from a dorm room at MIT, Cambridge, Massachusetts Acquired: July 2013 by Yahoo! for $60M+ Total VC Raised: ~$40 Million Outcome: Acqui-hired by Yahoo!, technology absorbed into Yahoo Mail; shut down one year later
The One-Line Summary
Two engineers, one from MIT, one from Penn State, met at a Washington D.C. internship, recognised that email was a broken mess of buried relationships, and built a product from a dorm room that Bill Gates personally demoed at a Microsoft conference. Then they turned down $20M from that same Microsoft, and eventually sold for $60M+ to Yahoo instead.
Lens 1, The Before State
Who Were They Before YC?
Adam Smith was a computer science student at MIT with a specialisation in machine learning and large-scale systems. He had spent years thinking about how information inside computers could be organised intelligently, not just stored, but understood. He was the kind of engineer who saw data as relationships, not just rows.
At age 13, he had already started a shareware company called Viratech, selling two products, Time Manager and EBMailer, for $20 each. Both sold well for a decade. By the time he got to MIT, entrepreneurship wasn't a dream. It was a habit he'd had since middle school.
Matt Brezina was a Penn State electrical engineering graduate, class of 2003. He came from a different world to Adam, a large state school, an engineering background, a mid-career internship track rather than a startup-native environment.
The two met while both interning in Washington D.C., not at a startup, not at a tech accelerator, but at ordinary internship jobs. They were not Silicon Valley insiders. They had no networks in venture capital, no warm introductions, no famous advisors. They were two smart engineers who happened to end up in the same city at the same time.
The Personal Pain They Were Living
The frustration was specific, and both of them had it. Email was the place where every important professional relationship lived, but the inbox had no idea who anyone was. You could have exchanged 400 emails with someone over three years, and when their message arrived, your email client treated them like a stranger. No context. No history surfaced. No relationship intelligence at all.
Adam put it clearly: "Not all email is created equal. Some messages may be relevant for years, while others lose meaning within minutes. Yet email inboxes treat all messages alike, regardless of who wrote them, what they're about, or when they were sent."
The inbox was the most important software tool that hundreds of millions of people used every single day, and it was essentially dumb. It stored things. It didn't understand them.
Why They Didn't Look Like "YC Founders" in 2006
In 2006, YC was only one year old. Most applicants were young, male, technical, and from well-known universities, so Adam and Matt technically fit a narrow profile. But they had no prior exits, no famous connections, no investors already circling them, and a product idea that sounded almost too boring: a plugin for Microsoft Outlook.
"Outlook plugin" is not a sentence that makes investors lean forward. Outlook was already built by the most powerful software company in the world. Most people in 2006 would have told you that trying to improve Microsoft's own email client was either redundant or suicidal. The conventional wisdom said: don't build on platforms you don't own.
Adam and Matt bet that conventional wisdom was wrong.
Key Insight for Aspiring Founders
The most boring-sounding markets often have the most pain. Nobody was excited about an Outlook plugin in 2006, but hundreds of millions of people used Outlook every single day and suffered through it. Boredom of category does not equal absence of pain. Find the frustration hiding inside the mundane.
Lens 2, The Idea Origin
How the Idea Was Actually Born
Adam Smith's core intellectual obsession at MIT was machine learning and how it could be applied to personal information. He looked at email differently from most people: he saw it not as a messaging tool but as a social graph, a rich, constantly updated record of every relationship a person had, hiding in plain sight.
Every email you sent or received contained data: who you communicate with most, how quickly you reply to certain people, what topics you discuss with whom, who introduces you to whom. All of that intelligence was sitting dormant inside Outlook. Nobody was reading it. Nobody was surfacing it for the user.
Adam's insight was straightforward: what if your email client knew who mattered to you, and showed you that context automatically?
The idea he and Matt built toward, which they eventually called Xobni Insight, would mine your email history and create a real-time social profile for every person you interacted with. When a message arrived from someone, you'd see their full contact history, shared attachments, threaded conversations, response times, and mutual contacts, all in a sidebar, without you having to search for anything.
The First Version, Built in a Dorm Room
The first version of Xobni was written in Adam's MIT dorm room in Cambridge, Massachusetts in early 2006. It was an Outlook plugin, meaning it had to be installed on a Windows PC, lived inside someone else's software, required no app store, no mobile, no cloud, and no elegant consumer onboarding. Just a .exe file, an install process, and the hope that people would actually run it.
This was deeply unfashionable. The tech world in 2006 was falling in love with web apps and Ajax. Outlook was "enterprise." Desktop plugins were "legacy." And yet, the decision to go deep on Outlook rather than build a light web tool was precisely what gave Xobni its analytical power. You can't mine 10 years of someone's email history from a browser tab. You need to be installed locally, close to the data.
The Signal That It Was Real
YC's Jessica Livingston wrote about Xobni at launch: "We funded them on the premise that they were going to 'make email better.' They've been working in stealth mode since then, but it seems like the Xobni guys will deliver."
That phrase, make email better, was the entire pitch to YC. Three words. The simplest possible framing of a deeply technical product. It worked because YC partners understood that everyone in the world had email and everyone found it frustrating. No market size calculation needed.
The Pattern This Follows
Xobni fits a specific YC archetype: extract intelligence from data people already have but can't use. The data existed, millions of emails, years of communication history, but no software was reading it on the user's behalf. This is the same insight that would later power LinkedIn, CRM tools, and eventually AI email assistants. Xobni was early by a decade. That timing is worth studying.
Lens 3, The Application Anatomy
Their YC Pitch, Three Words That Won
The pitch that got Xobni funded at YC in 2006 was functionally: "make email better."
That's it. Not a 10-slide deck. Not a TAM calculation. Not a competitive matrix against Microsoft. Three words that were true, relevant to everyone, and impossible to argue with. Every YC partner used email. Every YC partner was frustrated by it. The pitch landed because it required zero imagination to feel the problem.
For aspiring founders, this is a master class in clarity. A three-word pitch only works if the problem is universally felt. Xobni had that. If your problem requires long explanation, you haven't found the right framing yet.
What They Had at Application Time
| Factor | Xobni's Reality |
|---|---|
| Revenue | None |
| Paying customers | None |
| Working product | Early prototype / proof of concept |
| Prior exits | None |
| Academic credentials | MIT (Adam), Penn State (Matt) |
| Unique edge | Machine learning + email data insight |
| Category risk | High, building on Microsoft's platform |
You've read your 5 free stories this month.
The next YC application you write could be the one that gets in. Don't stop learning from the founders who already did it.
- ✓ 1000+ deeply-researched YC founder stories
- ✓ Unfiltered Lens breakdowns: what worked, what failed
- ✓ 2 new founder deep-dives every week
- ✓ Full Q&A library + application teardowns
Free YC databases
Everything behind this page is in our open databases
This story is one slice of the data we keep open — batch lists, rejection case studies and launch playbooks, all free to read.
Database
YC Rejection Database →
Airbnb, Stripe, Reddit — founders who got rejected first, with every source linked.
Index
List of YC Companies →
Searchable index of YC alumni by batch, from Airbnb (W09) to the newest AI startups.
Database
YC Solo Founder Database →
20+ verified solo founders, their pre-YC traction and the exact application framing they used.
Browse all free YC databases → · or start on the YCInsight homepage
Go deeper on what Xobni (pronounced "ZOB-nee" — "inbox" spelled backwards) did
Questions this story raises
- YC RFS Developer Tools — Specific Opportunities YC Wants Founders to Build
- What Makes a YC Application Stand Out From 50,000 Others
- What YC Partners Think About During Interviews — Insider View
- How Indian Founders Should Frame Their YC Application
- YC Interview Questions About Your Cofounder Relationship
- What Is YC Demo Day? Format, Investors, and What Actually Happens
Free tools for this stage
Start from the YCInsight homepage for every YC database, founder story and Q&A in one place.
Related founder stories
Keep reading — more YC founders from Summer 2006 and beyond.
Thinkature Founder Story: How Drew Harry Built Thinkature (Summer 2006 YC Batch)
Drew Harry built a real-time online whiteboard in 2006 that worked — got into YC, got press, got users — and still shut down four years later. The lesson isn't about the product. It's about building something people admire vs. building something people pay for.
Vidyard
Two engineering students in Canada — one designing toilets, one sitting idle at BlackBerry — used a day-trading windfall to buy a house, build a video company nobody asked for, and accidentally discovered the real product while trying to solve their clients' confusion. Then they drove 1,200 miles with a car covered in stickers to crash a conference and corner a keynote speaker.
Treehouse
Ryan Carson watched developers graduate from college without knowing how to code for real jobs — so he spent 6 years building a blog audience first, then launched a product to that audience and hit $1.7M revenue in under 12 months, teaching over a million people to code before anyone in Silicon Valley thought online education could work.
inDinero
A 19-year-old CS student who had never held a real job, knew nothing about accounting, and dropped out of high school at 15 — built the "Mint.com for business," nearly destroyed it through arrogance, then pivoted her way to 2,686% revenue growth and the cover of Inc. magazine.
RentHop
Two MIT-trained quants looked at Craigslist — the most-visited, least-innovated website in America — and decided the apartment hunt didn't need disruption with venture-scale fireworks. It needed a better ranking algorithm, patient bootstrapping, and fifteen years of quietly outlasting better-funded competitors.
Beetailer
A self-taught Spanish developer from Seville moved to San Francisco with a brilliant idea at exactly the right moment — got into YC, hit real traction, mentored by legends — and then watched everything collapse when Facebook changed the rules overnight. What she built next was smarter, and what she learned is priceless.
More stories every week.
Get 2 free stories in your inbox each week.