The founders describe a multiplayer game where beginners write code to move characters and defeat opponents, learning harder concepts as they progress. Their launch drew enough traffic from Reddit to overwhelm their server, reaching a reported 15,000 people. Graham probes how the team will create more levels and validate its recruitment business model; the founders say they are still testing whether companies will pay for qualified developer leads.
Office Hours with Paul Graham and Sam Altman
Speakers
At Y Combinator’s Startup School 2013 office hours, Paul Graham and Sam Altman discuss three startups with their founders: a multiplayer coding game, a daily video progress site, and Flexport, a web-based U.S. customs brokerage.
The Give It 100 founders describe a site where people post short daily videos while learning a skill. Its private beta had 50 users and a waiting list of about 4,300; examples included dancing, beatboxing, coding, and learning to walk. Altman pushes the team to test whether people will share videos on YouTube, while the founders point to their creator’s video, which had reached 3 million views. Graham advises them to observe which uses take hold before optimizing for one category.
The Flexport founder explains that the company combines a licensed customs brokerage with a web app that organizes documents and helps staff file import paperwork electronically. He says the team has three customers, a waiting list of about 300, and expects to charge around $100 per clearance, with extra fees for some shipments. The first clearance was delayed by a government shutdown, and the founder sees brokerage as a way into broader logistics services.
The Flexport discussion turns to how the company could expand from customs clearance into freight, warehousing, inspection, and trucking. The founder estimates that capturing 1% of the customs clearance business could yield about $30 million in annual profit, while acknowledging uncertainty in the figures. Graham questions whether very large market claims are persuasive, and the founder considers sales teams as one way to reach importers.
Takeaways
- Design learning around meaningful goals and let difficulty rise with the learner’s skills.
- Treat launch traffic as a test of product demand and infrastructure readiness.
- Test the specific sharing behavior your growth plan depends on.
- Watch which customer uses emerge before committing the product to one category.
- Use a focused initial service to build customer knowledge before expanding into adjacent offerings.
Mentioned in this talk
Search inside this talk
What was claimed 8
- featureGive It 100: Users choose a skill or goal and share a progress video each day
“share a video of your progress every day”
YouTube Unverified1 source· Talk
- metricGive It 100: 50 invite-only beta users and about 4,300 people on the waiting list
“50 people, and there's about 4,300 on a waiting list”
YouTube Unverified1 source· Talk
- metricGive It 100: About 1,000 video views per day
“about a thousand views a day”
YouTube Unverified1 source· Talk
- usesGive It 100 uses
YouTube
“We want to use YouTube”
YouTube Unverified1 source· Talk
- featureFlexport: Licensed U.S. customs brokerage using a web dashboard to organize import documents and file customs entries electronically
“the first licensed U.S. customs brokerage built around a modern web application”
YouTube Unverified1 source· Talk
- metricFlexport: Three customers importing goods and a waiting list of about 300
“we have three customers importing stuff. We actually have a waiting list of about 300”
YouTube Unverified1 source· Talk
- metricFlexport: Estimated gross margin of about 75% per clearance
“the gross margin should be about 75%”
YouTube Unverified1 source· Talk
- metricFlexport: Plans to charge $100 per clearance, with possible extra charges for additional paperwork
“We're going to charge $100 to do it”
YouTube Unverified1 source· Talk






