India Can Create The Largest AI Companies

Predictions and Expectations:

• Puneet expects "some of the largest companies in the world will come out of india" because of its technical talent.
• Puneet predicts that "AI change is global" and that "this is our time to shine" for India, contrasting it with the previous local mobile revolution.
• Puneet believes it is now "unimaginable" for a cold email from an Indian student to sell to US insurance companies, citing a specific recent example.
• Arnav expects the future of the India AI ecosystem to be "defined by all the people in this room and not by people who came from a prior generation."
• Arnav believes that "a lot of those jobs [banker, consultant, engineer, doctor] won't exist 10 years from now, or at least they will change dramatically."
• Arnav suggests the "safe path" of traditional employment might be the "risky path" if one believes in the long-term trajectory of AI.
• Arnav states that people with "high agency" are favored by the current AI era, whereas non-AI native people giving traditional advice are "very dangerous" to follow.
• John predicts that AI has "completely leveled the playing field for young founders" regarding the ability to build and the pace of learning.
• John observes that "almost every founder had the experience of their big idea was not their first idea" and notes that many successful ideas are not first-mover concepts.
• John expects that using coding agents allows people to "make their ideas become real extremely fast" and often with more quality than expected.
• John believes that the "second mover effect" is strong because competitors can be beaten with a better product unless the first mover has "incredible network effects."
• John predicts the "cost of compute" and "level of intelligence" is going to "go down a lot."
• John suggests that companies wanting to give employees "unlimited budgets" to test models at the frontier will become more common.
• John expects that being "early in the game" and trying to build for the model "six months ahead, a year ahead" will provide a huge competitive advantage.
• John believes that the fundamental truths of what makes a great founder (obsessed with customers, tinkering, building at the edge) "have changed weirdly surprisingly little" and likely "won't change in 20 years."
• Arnav hopes that the attendees will "consider that [starting a company] as well" as an alternative to the well-trodden path.
• Arnav expects that "teams form from here or from your best friends back home" over the next few days to use the provided credits.

Timelines and Milestones:

• Arnav notes that the "AI boom" is currently only "year three or four."
• Arnav estimates that the attendees have "easily 30 40 years left to go build."
• John mentions that the average age of YC founders has gotten younger over the "last sort of 10 20 batches."
• John notes that he started paying for the top-tier coding model "December of this last year" or during "this christmas break."
• Arnav states that attendees will receive an email with credits "tonight or tomorrow" and can redeem them "right away" to get them "within a few days."

Technology and Product Direction:

• Puneet advises founders to focus on "living at the edge of the technology" and understanding it "10x better than everyone else" rather than focusing on go-to-market or business models.
• Arnav suggests that founders should "build for the model six months ahead, a year ahead" to stay ahead of competition.
• Arnav predicts that "open source models will keep getting better and better" while token costs "keep coming down."
• John expects that coding agents allow founders to "beat benchmarks," "establish new heights," and "figure out where the bottlenecks are" by tinkering.
• John suggests that founders should "find something that's kind of working and then do it better than them" rather than always being the first mover.
• John advises spending "tokens" extensively (e.g., writing "10 000 unit tests") rather than being bandwidth constrained by cost to unlock better code.
• Ankit defines a "project" as two people building something unassigned, getting it used, and notes that doing "projects especially when you're young" guarantees finding startup ideas.

Market and Industry Outlook:

• Puneet believes that "anyone anywhere could say that I have one hour free" due to the mobile revolution, but the AI revolution is not local, leading to hyper-local network effects previously and now global opportunities.
• Arnav observes that the "Indian startup ecosystem is booming, but maybe not as deep as the US."
• Arnav notes that in the US, it is "expected that you're going to be ambitious," whereas in the Indian education system, "it's not cool to be ambitious."
• Puneet predicts that "selling to insurance company in the us buy a cold email" is now a viable strategy for Indian founders without connections.
• John notes that the best young founders are "tinkering more" and "following their curiosities" to work on the edge of what is possible.

Company Plans:

• Puneet states that he is "back to building" after his investor stint.
• Arnav notes that he is "mostly trying to work with you know all the young builders building an ai and investing in them."
• Six companies (including Giga, Emergent, Zepto, etc.) are "very actively hiring engineers" and want people to "go apply on their website."
• The organizers plan to "hand out swag bags" and have "food and photos on the lawn" to encourage networking.
• The organizers plan to email attendees "tonight or tomorrow" giving access to credits and encourage them to start building "amazing things."

Financial Guidance:

• Puneet mentions that Gary Tan (Giga) "spends thousands of dollars per day on token usage."
• John reveals that the Anthropic "max plan" is "$200 a month" and implies this is the threshold to reach the frontier for coding.
• John suggests that if a founder is willing to spend "$5 per email" on inference, the output quality for email automation would be "really good."
• Ankit offers to give away "the largest credit packages" to whoever gets a product off the ground the fastest using the provided credits.

Risks and Caveats:

• Arnav cautions that it is "very hard for an average young Indian to take the types of risks that someone who grew up in Silicon Valley can, just given the social safety nets."
• Arnav warns that following the advice of people who are "not ai native" is "very dangerous."
• Arnav acknowledges that a "safe, stable, prestigious" job is already "the top 10% of this country" for many, so it is not a bad path for everyone.
• Puneet notes that the previous mobile revolution led to "hyper-local network effects" and implies the AI revolution is different in scope.
• John notes that most founders go through "some number of pivots" and that "a good startup idea is not obvious."
• John warns that unless a first mover has "incredible network effects," a second mover can beat them with a better product, implying network effects are a risk for new entrants.

Confidence and Disagreement:

• Puneet is "very, very excited" about the ability to build global companies out of India.
• Arnav is "willing to bet there are a lot of people in this room that disagree with that cookie cutter advice."
• John is confident that the "cost of compute... is going to go down a lot."
• John is certain that the fundamental truths of great founders "won't change in 20 years" because they have held true "since Thomas Edison."
• Ankit is confident that doing projects will make one "guaranteed to find startup ideas."
• Arnav states that he is "not exactly saying this is my advice" regarding the second-mover strategy, presenting it as a "reasonable way to compete right now."
• Arnav is "of two minds" about open source models, seeing them as useful for mass adoption but less so for the "very frontier" of tasks.