{"id":15193,"date":"2026-08-22T05:24:41","date_gmt":"2026-08-22T05:24:41","guid":{"rendered":"https:\/\/koshalsambada.in\/?p=15193"},"modified":"2026-08-22T05:24:41","modified_gmt":"2026-08-22T05:24:41","slug":"y-combinator-ceo-has-a-message-for-all-the-founders-and-ceos-worried-about-burning-too-many-ai-tokens","status":"publish","type":"post","link":"https:\/\/koshalsambada.in\/?p=15193","title":{"rendered":"Y Combinator CEO has a message for all the founders and CEOs worried about burning too many AI tokens"},"content":{"rendered":"<p><br \/>\n<\/p>\n<div>\n<div class=\"e9jwa\">\n<div class=\"vdo_embedd\">\n<div class=\"GfdvZ\">\n<section class=\"_bIDB  clearfix id-r-component leadmedia undefined undefined  E9tg9 \" style=\"top:0px\">\n<div class=\"_bIDB\" data-ua-type=\"1\" onclick=\"stpPgtnAndPrvntDefault(event)\">\n<div class=\"ypVvZ\">\n<div class=\"WGttI\"><img src=\"https:\/\/static.toiimg.com\/thumb\/msid-133417694,imgsize-624744,width-400,height-225,resizemode-4\/y-combinator-ceo-garry-tan.jpg\" alt=\"Y Combinator CEO has a message for all the founders and CEOs worried about burning too many AI tokens\" decoding=\"async\" fetchpriority=\"high\"\/><\/div>\n<\/div>\n<\/div>\n<\/section>\n<\/div><\/div>\n<\/div>\n<p>\u200b<span class=\"id-r-component br\" data-pos=\"1\"\/>Y Combinator CEO Garry Tan has a message for founders and CEOs concerned about the cost of using large numbers of AI tokens: \u201cBurn, baby, burn.\u201d Speaking on an episode of the A16z podcast this week, Tan argued that founders should spend heavily on AI agents, even when doing so can result in high annual costs.<span class=\"id-r-component br\" data-pos=\"4\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"6\"\/>Tan said founders should give themselves permission to use large token budgets when working with AI agents, arguing that doing so can provide access to a level of AI capability that he believes will otherwise become available only in the future.<span class=\"id-r-component br\" data-pos=\"8\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"10\"\/>\u201cYou have to tune it all the way up. You&#8217;re just, like, let me load a million tokens or 800,000 tokens in,\u201d he said about using AI agents. \u201cWhen you do that, I think that you basically get to live in 2028.\u201d<span class=\"id-r-component br\" data-pos=\"14\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"16\"\/><\/p>\n<p><h2>Garry Tan backs heavy AI token spending<\/h2>\n<\/p>\n<p><h2>\u200b<\/h2>\n<\/p>\n<p>Tan acknowledged that using AI agents at full capacity can be expensive. He estimated that the cost could reach tens of thousands of dollars per year, but said the spending can make sense for founders and CEOs experimenting with AI-driven workflows.<span class=\"id-r-component br\" data-pos=\"20\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"22\"\/>\u201cIt costs, I don&#8217;t know, 50 or $100,000 a year to use the agents at full strength,\u201d he said. \u201cIt just costs like a crazy amount. But for a CEO or for a founder, it actually makes a lot of sense to do that.\u201d<span class=\"id-r-component br\" data-pos=\"25\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"27\"\/>His approach is to use AI agents for tasks first, then turn successful workflows into repeatable processes. Tan said founders can document the steps an agent follows and use those instructions to automate the same work in the future.<span class=\"id-r-component br\" data-pos=\"29\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"31\"\/>\u201cA markdown file is an employee,\u201d he said, referring to a written set of instructions that tells an AI agent how to perform a particular job. \u201cIt&#8217;s an employee that will do the job perfectly every single time, and it&#8217;ll do it as many times as you want.\u201d<span class=\"id-r-component br\" data-pos=\"34\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"36\"\/><\/p>\n<p><h2>AI <keyword id=\"34406385\" type=\"Denny\" weightage=\"20\" keywordseo=\"tokenmaxxing\" source=\"keywords\">tokenmaxxing<\/keyword> divides Silicon Valley<\/h2>\n<\/p>\n<p>\u200b<span class=\"id-r-component br\" data-pos=\"39\"\/>Tan&#8217;s approach contrasts with criticism of \u201ctokenmaxxing,\u201d a term used for deliberately allowing AI systems or agents to consume large numbers of tokens instead of prioritising lower usage and costs.<span class=\"id-r-component br\" data-pos=\"41\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"43\"\/>Uber&#8217;s tech chief Praveen Neppalli Naga recently argued that companies may be moving away from this approach as they focus more closely on the efficiency of AI spending.<span class=\"id-r-component br\" data-pos=\"45\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"47\"\/>\u201cThe next phase, whatever we call it, will not be characterised by who spends the most tokens, but about how people use them as efficiently as possible,\u201d he said in a post on X.<span class=\"id-r-component br\" data-pos=\"50\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"52\"\/>Cognition CEO Scott Wu also questioned the way some companies have approached tokenmaxxing. Speaking on a podcast in June, Wu said companies could lose sight of the relationship between token consumption and actual work produced.<span class=\"id-r-component br\" data-pos=\"54\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"56\"\/>\u201cIt is directionally correct, but I think there are definitely some places where people have gotten carried away,\u201d Wu said. \u201cPeople are like, &#8216;We rank our engineers by how many tokens they&#8217;re spending.&#8217; <!-- -->Well, let&#8217;s try and rank people by how much output they&#8217;re actually producing.\u201d<span class=\"id-r-component br\" data-pos=\"61\"\/><\/p>\n<p><h2>\u200b<\/h2>\n<\/p>\n<p><h2>AI agents become part of founder workflows<\/h2>\n<\/p>\n<p>\u200b<span class=\"id-r-component br\" data-pos=\"65\"\/>Tan&#8217;s argument focuses on using higher AI token limits as an investment in experimentation rather than treating token consumption only as a cost to minimise. He recommends first allowing agents to use the resources needed to complete a task, then converting successful processes into reusable instructions.<span class=\"id-r-component br\" data-pos=\"67\"\/>\u200b<span class=\"id-r-component br\" data-pos=\"69\"\/>The approach reflects a broader discussion among technology companies about how much they should spend on AI inference as increasingly capable models and agents are used for software development, research and other business tasks.<span class=\"id-r-component br\" data-pos=\"71\"\/>\u200b<\/div>\n<p><br \/>\n<br \/><a href=\"https:\/\/timesofindia.indiatimes.com\/technology\/tech-news\/y-combinator-ceo-has-a-message-for-all-the-founders-and-ceos-worried-about-burning-too-many-ai-tokens\/articleshow\/133417589.cms\" target=\"_blank\" rel=\"noopener\">Source link <\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u200bY Combinator CEO Garry Tan has a message for founders and CEOs concerned about the cost of using large numbers of AI tokens: \u201cBurn, baby, burn.\u201d Speaking on an episode of the A16z podcast this week, Tan argued that founders should spend heavily on AI agents, even when doing so can result in high annual [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":15194,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[31],"tags":[],"class_list":["post-15193","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-31"],"magazineBlocksPostFeaturedMedia":{"thumbnail":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","medium":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","medium_large":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","large":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","1536x1536":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","2048x2048":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","blogsy-small":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","blogsy-small-tall":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","blogsy-small-square":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","blogsy-small-masonry":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","blogsy-medium":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","blogsy-medium-masonry":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","blogsy-large":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg","blogsy-wide":"https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg"},"magazineBlocksPostAuthor":{"name":"admin","avatar":"https:\/\/secure.gravatar.com\/avatar\/8709732a479614e7a8aa24d3eb1b239f30dc6d90c61464ed495001e7a469d856?s=96&d=mm&r=g"},"magazineBlocksPostCommentsNumber":"0","magazineBlocksPostExcerpt":"\u200bY Combinator CEO Garry Tan has a message for founders and CEOs concerned about the cost of using large numbers of AI tokens: \u201cBurn, baby, burn.\u201d Speaking on an episode of the A16z podcast this week, Tan argued that founders should spend heavily on AI agents, even when doing so can result in high annual [&hellip;]","magazineBlocksPostCategories":["\u0b26\u0b47\u0b36 \u0b2c\u0b3f\u0b26\u0b47\u0b36"],"magazineBlocksPostViewCount":2,"magazineBlocksPostReadTime":3,"magazine_blocks_featured_image_url":{"full":["https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg",400,225,false],"medium":["https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg",300,169,false],"thumbnail":["https:\/\/koshalsambada.in\/wp-content\/uploads\/2026\/08\/y-combinator-ceo-garry-tan.jpg",150,84,false]},"magazine_blocks_author":{"display_name":"admin","author_link":"https:\/\/koshalsambada.in\/author\/admin"},"magazine_blocks_comment":0,"magazine_blocks_author_image":"https:\/\/secure.gravatar.com\/avatar\/8709732a479614e7a8aa24d3eb1b239f30dc6d90c61464ed495001e7a469d856?s=96&d=mm&r=g","magazine_blocks_category":"<a href=\"#\" class=\"category-link category-link-31\">\u0b26\u0b47\u0b36 \u0b2c\u0b3f\u0b26\u0b47\u0b36<\/a>","_links":{"self":[{"href":"https:\/\/koshalsambada.in\/index.php?rest_route=\/wp\/v2\/posts\/15193","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/koshalsambada.in\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/koshalsambada.in\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/koshalsambada.in\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/koshalsambada.in\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=15193"}],"version-history":[{"count":0,"href":"https:\/\/koshalsambada.in\/index.php?rest_route=\/wp\/v2\/posts\/15193\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/koshalsambada.in\/index.php?rest_route=\/wp\/v2\/media\/15194"}],"wp:attachment":[{"href":"https:\/\/koshalsambada.in\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=15193"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/koshalsambada.in\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=15193"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/koshalsambada.in\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=15193"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}