Punit grew Super Daily to an annual run rate of $100 million ARR with one engineer (two including himself) and sold it to Swiggy. This happened in the pre-AI era.
Link to this point #1Y Combinator
India Can Create The Largest AI Companies
Watch the original on YouTube ↗AI-generated summary. Check the original video for context and accuracy. Translated into English with AI.
Overview
The final panel discussion of Startup School India, hosted by Y Combinator. YC partners and former YC alumni investors discuss the potential for India's excellent technical talent to create some of the world's largest companies in the age of AI. They also talk about how young founders can grow beyond the advice of traditional education systems.
Key points
Unlike the mobile revolution, the AI revolution is global, not local. It's not about hyperlocal network effects, but about being at the edge of technology, and India, with its best technical talent, can create the world's largest companies.
Link to this point #2In the latest YC batch, an Indian team cold-emailed US insurance companies and secured contracts. It's now an era where IIT 3rd-year students can succeed in the US market.
Link to this point #3The average age of YC founders has gotten younger over the last 10-20 batches. AI has removed the limitation of building ability, making learning speed the only limiting factor, which benefits young founders.
Link to this point #4High-paying, stable professions traditionally recommended by the education system (bankers, consultants, engineers, doctors) may not exist or change dramatically in 10 years, making historically safe paths potentially risky now.
Link to this point #5Good startup ideas are often not the first ones. Almost all founders who spoke today experienced multiple pivots, realizing truly good ideas during the actual building process.
Link to this point #6Varun, the founder of Giga, competed against rivals with hundreds of employees and secured a contract with Door Dash with an 8-person team, demonstrating that strong technical teams can win even with a small number of people.
Link to this point #7Gary Tan spends thousands of dollars per day on token usage. Unless you subscribe to the $200/month max plan, you cannot access the true performance of current frontier models.
Link to this point #8The definition of a project is 'two or more people building something not assigned to them and getting someone to use it.' You can be successful in computer science education or jobs and never do a project by this definition.
Link to this point #9YC is holding essentially the same event as Startup School in 2013. Since Thomas Edison, the basic characteristics of great founders (customer obsession, experimentation, and building at the edge of technology) have not changed.
Link to this point #10Put this knowledge to work.
Use saved videos in Claude Code or Codex to think through your plans and code. Sign up and connect your AI to get started.