Daily Brief

Starting a company is hard. Reinventing your company for AI as a public company with quarterly earnings results is even harder. Aaron Levie has pulled off the transition with Box and offers hard-won advice for founders. The cofounder and CEO of Box argues the value isn't only in the model; it's in the bridge from a model's raw capability to the actual workflow inside a bank, a law firm, or a pharma company. That's the case for the application layer, and Box is building it: an agent harness tuned so tightly to its own file system, permissions, and search that it beats handing the raw API to Cla

Nearly every intelligence gain in language models over the past year has come from letting them think longer. Voice agents have to turn thinking off, because the budget between a user finishing a sentence and the agent starting to speak is measured in hundreds of milliseconds. Venky B is founder and CEO of Plivo, which carries over a billion voice calls a month and has been building telephony infrastructure since 2011, and this talk is a tour of what breaks when a voice agent leaves the demo and meets production. On latency his numbers are blunt. Teams aim for under 550 milliseconds and most l

A background agent quietly makes the tool call, then drops the result into the main model's context so the model believes it made the call itself. That sleight of hand is one of three tricks Bohan Li uses to run a voice agent on a slow, genuinely intelligent model without the caller noticing the wait. Li works on the voice harness at EliseAI and came from self driving, and he maps the cascaded voice stack onto that world: transcription is perception, the language model is planning, and speech synthesis is control. Each layer then gets optimized on its own terms. For transcription he runs two e

Midam Kim spells her first name for a voice agent, letter by letter. The bot reads it back with an N on the end. She corrects it. The bot replies thank you for your correction, then says her name wrong anyway, because its pronunciation rules are English and her name is not. Moments later it asks for an account number she has to go looking for, cuts her off while she reads the unfamiliar string aloud, announces it cannot find her record, and asks her to repeat the whole thing without trying anything different. She asks for a human. Kim is an ML engineer at ServiceNow and a longtime researcher o

Amit Desai never makes the model better. Accuracy stays pinned at 79 percent for the whole talk, and user pain still drops by nearly half. Desai has worked on voice interfaces for years, at Roku and before that on a widely used smart speaker, and his argument is that teams pour effort into one knob while ignoring a second one that is entirely independent of it. Knob one is accuracy, chased a percentage point at a time through the wake word, recognition, and understanding layers. Knob two is what the system decides to do when it is unsure. Take a thousand music requests where 790 play the right

Full episode:https://youtu.be/PrSf7IOYu-I Me on twitter: https://x.com/dwarkesh_sp

Ask your own company what it actually contracted for with a given vendor and nobody can tell you. The number exists, negotiated carefully, sitting in a pricing table inside a signed PDF no system ever read back. Hiral Shah puts a figure on the aggregate: roughly two trillion dollars of negotiated value locked in agreements that organizations never return to, because recovering it means human reading, human review, and manual work across disconnected systems. The scale on Docusign's side is its own engineering problem. Close to two million paying customers, a billion users, and about a million

Legora's search latency once went from a 100 millisecond P99 to twenty seconds, and the cause was a packing problem invisible in the schema. Legal work arrives as projects, a project gets worked hard for a while and then is never opened again, and when the team sharded document chunks across four thousand partitions the dead projects landed in the same partitions as the live ones. Every query pulled an enormous partition into memory, evicted the last, and thrashed the cache. Jacob Lauritzen walks the whole migration honestly, from one search cluster for everybody, then one per region once clie

Somewhere between 30 and 50 percent of an agent's tokens get spent on searching, almost all of it up front, before any of the work you asked for. Maximilian David Rumpf treats that number as the whole opportunity. Handing a task to an agent rather than a query to a search engine roughly doubles the odds of finding the right documents, but it costs a hundred to a thousand times more and takes minutes instead of milliseconds. His diagnosis of why the classical alternative cannot close that gap is the sharpest part. A traditional pipeline rewrites the query, hits a backend, reranks, and returns,

Ask a corpus of Seinfeld transcripts for the name of Jerry's favorite church and a 100 token chunk returns it at rank one, while every larger window buries it below rank fifty. Ask the same corpus who Jerry calls his nemesis and pure evil and the small chunks fail completely, because the answer is spread across a scene rather than sitting in a sentence. Same data, same index, opposite requirements. Yuval Belfer uses that pair to make a claim most retrieval teams have quietly assumed away: there is no correct chunk size, because the correct size is a property of the query, and you pick it at in

Full episode:https://youtu.be/PrSf7IOYu-I Me on twitter: https://x.com/dwarkesh_sp

From their seed led by Nat Friedman and Dan Gross, AIUC has now announced a monster $40M Series A! From being Anthropic’s first product hire to building the standards, testing, and insurance infrastructure meant to make frontier AI deployable, Rune Kvist is betting that the biggest constraint on AI adoption won’t be capability — it will be trust. In this episode, the AIUC cofounder joins swyx and Vibhu to announce a new $40M round and explain why companies like Cursor, Harvey, Lovable, and ElevenLabs are increasingly confronting a problem that gets harder as AI gets better: who is responsible

教师节，祝老师们节日快乐，也感谢那些给我托举，给我祝福，给我无私帮助的师长。

本篇为世纪风暴系列伊朗篇——告别伊玛目。

法治建设，它不可能一蹴而就，无法毕其功于一役，法治的每一个进步背后都有许多沉甸甸的故事，这些故事不应该被人遗忘，因为牢记历史就是为了避免重蹈覆辙。 参考文献： 1.陈永生，我国刑事误判问题透视———以20起震惊全国的刑事冤案为样本的分析，中国法学2007年第3期 2.对我国死刑复核程序之检讨———以中国古代及国外的死刑救济制度为视角，陈永生，比较法研究2004年第4期。