From application to demo
118 builders applied. 50 stood on stage with something they'd shipped.
The recap
The Founder Sprint — July 2026
The first Founder Sprint: 48 hours, 100 hand-picked builders, one validation methodology.
The first Founder Sprint brought 100 hand-picked builders to Bellevue for 48 hours. Not to ship the most features in a weekend — to find the strongest evidence that their idea actually matters to real people.
Teams left with a working prototype, real signal from the market, and a business thesis they could actually defend. Here is how it went.
48
HRS
Of building
100
BUILDERS
Hand-picked
2
DAYS
July 11–12
$23K+
PRIZES
Cash + credits
Moments from the room
Launch Day, by the numbers
Forty-eight hours, measured. The honest story of what twenty-five teams built — and where the signal actually was.
Who showed up
100 accepted builders — students, operators, and career-switchers.
- ~36% came from a non-CS domain
- 60/100 arrived seeking an idea — and most formed teams on-site
- 46% 32% solo share fell 14 points as builders paired up
What they built
25 teams, by sector.
18 of 25 teams built an AI agent to automate one specific job.
The scorecard
Building beat validating by 0.75 of a point
Field average, out of 5. The three build dimensions (purple) all out-scored the three validation dimensions (amber).
Everyone did the math. Few made the call.
Out of 25 decks.
Market-sizing was easier to reach for than customer evidence — only 6 of 25 teams cited a real conversation.
The riskiest assumption on the table
Each team's make-or-break bet — how many made it, and how they scored (avg /30).
- Behavior / adoption 11 teams avg 15.9/30
- Willingness to pay 6 teams avg 17.2/30
- Data / trust access 4 teams avg 15.5/30
- Technical feasibility 3 teams avg 17.8/30
- Defensibility 1 team avg 17.0/30
The most common bet — betting on behavior change — was the weakest-scoring.
The verdict
Who won the room
🏆 Winner · highest judge score
Soglia 22.8 / 30
AI that wins grant funding for small charities on a 5% success fee — the only team above 3.0 on all three validation dimensions.
- Crowd Favorite Pause AI
- Boldest Bet Khan · AI desalination
- Best Pitch Soglia
- Wildcard Khan
Signal over speed
Anyone can build an app in a weekend now — AI ships code in hours. The hard part was never the code; it was knowing whether anyone wanted what you built. So the Sprint rewarded evidence, not feature count: teams spent the weekend talking to real people, running real experiments, and following the signal.
A room full of people trying to do the same thing
Builders worked alongside operators, founders, and investors who had actually shipped. Mentors pressure-tested assumptions in real time, and the judging panel scored on the strength of the evidence and the clarity of the thesis — not the polish of the demo.
Judges & mentors
-
Taimur Rashid
-
Alaa Badr
-
Riham Mansour
-
Jonathan Chang
-
Tarek Elabbady
-
Usman Shakeel
The team behind it
-
Mohamed F. Ahmed
-
Azeez Idris
-
Ahmad ElSwify
The crew who made it run
Operators, students, and post-grads who gave their weekend so 100 builders could ship something real — 13 crew on board.
-
Syed Shayaan Ali
-
Zobir Tarin
-
Ikram Bennour
-
Zawad Ahmed
-
Daaniya Junejo
-
Monisa Ansha
-
Hafsa Abdinur
-
Shruthika Balasubramanian
-
Ayesha Khan
-
Habiba Elswify
-
Yousef Gomaa
-
Noor Badran
-
Mohamed Elazhary
The next Sprint is in October.
Want in for the October 2026 edition? Drop your name and email and you'll be the first to hear when applications open.