Search the archive

YOUR DATA MODEL IS YOUR DESTINY

A 'data model' is the core concept or object a startup prioritizes and builds around.

JIT: SO YOU WANT TO BE FASTER THAN AN INTERPRETER ON MODERN CPUS…

Modern CPUs have an insane level of optimization, but this isn't always helpful.

Register today →

OPENAI, BROADCOM FORGE MULTIBILLION-DOLLAR CHIP-DEVELOPMENT DEAL

OpenAI and Broadcom will develop and deploy 10 gigawatts of custom AI chips and computing systems over the next four years. The new agreement, worth multiple billions of dollars, brings the total scale of computing capacity OpenAI has agreed to buy to 26 gigawatts. OpenAI plans to deploy custom GPUs starting in the second half of next year. It is aiming to build 250 gigawatts of new computing capacity by 2033.

FACEBOOK'S JOB BOARD RETURNS AS AI THREATENS ENTRY-LEVEL WORK

Facebook is reviving its jobs board for US users on Android and iOS. The job listings will cater to young adults trying to find entry-level job opportunities. They will be public and visible to users 18 years and older. Users will be able to directly reach out to employees through Messenger to ask questions or schedule interviews.

STARSHIP'S ELEMENTARY ERA ENDS TODAY WITH MEGA-ROCKET'S 11TH TEST FLIGHT

Starship lifted off from Starbase at around 6 PM last night for its eleventh flight test. Every major objective of the flight test was achieved. Starship successfully deployed eight Starlink simulators and executed the third in-space relight of a Raptor engine during the flight. It also successfully executed a landing flip, landing burn, and soft splashdown in the Indian Ocean.

SUPERPOWERS FOR SUPERHEROES

Anduril's EagleEye puts mission command and AI directly into operators' helmets. It gives soldiers enhanced perception, lethal connectivity, and heightened survivability. The HUD overlays digital information onto the real world to deliver vital real-time insights. Anduril partnered with Meta, Oakley, Qualcomm, and Gentex to bring the best in AR, rugged eyewear, compute, sensing, and ballistic helmets to EagleEye.

AI Use Case Discovery Guide.

NANOCHAT

nanochat is a minimal full-stack training/inference pipeline for large language models (LLMs) designed to run on a single 8XH100 node. It can train an LLM in as little as four hours. nanochat handles tokenization, pretraining, finetuning, evaluation, inference, and web serving.

BEADS

Beads is a lightweight memory system for coding agents with a graph-based issue tracker. It helps agents be more organized and increases their ability to handle long tasks over multiple sessions. Beads enables agents to use issue tracking with proper epics, which makes it easier for them to reliably perform complex task streams in the right order. The issue tracker has a sophisticated audit trail, allowing agents to reconstruct complex operations.

DUTCH GOVERNMENT TAKES CONTROL OF CHINESE-OWNED CHIPMAKER NEXPERIA IN ‘HIGHLY EXCEPTIONAL' MOVE

The Dutch government has taken over Nexperia, a Chinese-owned semiconductor maker based in the Netherlands. It claims it invoked the 'Goods Availability Act' to prevent a situation in which the goods produced by the company would become unavailable in an emergency due to recent and acute signals of serious governance shortcomings and actions. Nexperia's daily operations will continue. The impact of the measures is not yet quantifiable.

WHY SIGNAL'S POST-QUANTUM MAKEOVER IS AN AMAZING ENGINEERING ACHIEVEMENT

Signal's latest updates are a significant step towards the messaging platform becoming fully quantum-resistant. Nothing will change for users as the changes are all under-the-hood - the apps will behave almost exactly as they did before. Signal has documented in great depth about the new design along with mathematical proof verifying that the updated protocol provides the claimed security properties. This article provides an overview of how the updated protocol works.

DEVREL IS -UNBELIEVABLY- BACK

Searches for 'developer relations' are at an all-time high.

JSONRIVER

jsonriver is a tool that can parse JSON incrementally as it streams in.

JPMORGAN JUST DRAFTED ITSELF INTO AMERICA'S INDUSTRIAL WAR

JPMorgan Chase has announced a $1.5 trillion plan that will mobilize capital for supply chains and advanced manufacturing, defense and aerospace, energy independence, and frontier technologies.

Subscribe now.

PG LINTER

PG Linter is a PostgreSQL extension that analyzes databases for potential issues, performance problems, and best practice violations.

FREE SOFTWARE HASN'T WON

Closed software is everywhere around us without free options.

AMPLIFIER

Amplifier is a development system that supercharges AI assistants with discovered patterns, specialized expertise, and powerful automation.

QA Wolf [5] takes testing off your team's plate by combining AI

with expert QA engineers to build and maintain 80% automated coverage [7] in weeks, not months. Tests run in parallel, suites finish in minutes, and failures are real—not false alarms. Workflows stay current as your app evolves, and brittle tests are repaired automatically.

releases ship faster with fewer surprises. Book a demo now.

INSIDE APPLE'S PIVOT FROM A ‘VISION AIR' HEADSET TO META-LIKE SMART GLASSES

Apple's shift to smart glasses suggests that the company no longer views the Vision Pro approach as a viable mass-market platform. The company doesn't invest heavily into products that lack the potential to become mainstream hits, and it doesn't chase niche markets. Smart glasses are a much more portable and practical form factor that could one day evolve into a future iPhone replacement, something the Vision Pro can never achieve. While Apple may someday decide to make the Vision Air after all, the decision now to focus on smart glasses is the logical path.

THINKING MACHINES LAB CO-FOUNDER DEPARTS FOR META

Andrew Tulloch, one of the co-founders of Mira Murati's Thinking Machines Lab, has left to join Meta. Tulloch previously worked at Meta for 11 years. He is a leading researcher in the field of artificial intelligence and worked at OpenAI before co-founding Thinking Machines Lab. It is unclear what team he will be on at Meta.

SCIENTISTS DISCOVERED A NEW CREATURE THAT EXISTS BETWEEN LIFE AND NOT-LIFE

Sukunaarchaeum mirabile is a cellular entity that offloads certain biological functions onto hosts like a virus, but it also contains the necessary genes to create its own ribosomes and messenger RNA. With just 238,000 base pairs of DNA, Sukunaarchaeum pushes the conventional boundaries of cellular life. Its discovery highlights the vast unexplored biological novelty within microbial interactions.

CHINA'S 'DARWIN MONKEY' IS THE WORLD'S LARGEST BRAIN-INSPIRED SUPERCOMPUTER

Darwin Monkey, also known as 'Wukong', is a supercomputer built to emulate a monkey's brain structure. It features over 2 billion artificial neurons and more than 100 billion synapses, roughly on par with the neural structure of a macaque. The supercomputer will serve as a tool for neuroscientists while also acting as a stepping stone toward artificial general intelligence. It consumes just 2,000 watts of power despite being powered by 960 Darwin III neuromorphic chips, each supporting up to 2.35 million spiking neurons.

VIBING A NON-TRIVIAL GHOSTTY FEATURE

Ghotty's unobtrusive update notification feature was largely developed with AI. This post shares the agentic coding sessions that led to shipping the feature. It provides context about the process and reasoning. AI is best used as an assistant and not a replacement. While AI can do a lot of the work, developers will almost always still need to intervene at times.

PEEKING INSIDE GIGANTIC ZIPS WITH ONLY KILOBYTES

WHAT IF SPORTSCENTER AND LINKEDIN MERGED?

TBPN (Technology Business Programming Network) is a show where Jordi Hays and John Coogan chat about the worlds of tech and business. The show became a sensation practically overnight after its first episode aired last October. Executives from OpenAI, Palantir, and Andreessen Horowitz appear regularly on the show. The show's audience is relatively small, but it considers CNBC and Fox Business to be its competitors. Its fans include Larry Ellison, Mark Cuban, and Travis Kalanick.

SELF ACTUALIZATION

When Andrew Bosworth started Meta's Bootcamp program in 2008, he didn't teach the culture as it was: he taught it as he wished it were. New engineers didn't know the difference. The company grew fast, and within a year, more than half of the engineering team had learned about the culture from him. When onboarding in a high-growth organization, you aren't just teaching people how things work, you're deciding what kind of company they believe they've joined.

PREDICTION: THE SUCCESSOR TO POSTMODERNISM

Prediction markets are perfectly suited for the zeitgeist right now.

SUPERPOWERS: HOW I'M USING CODING AGENTS IN OCTOBER 2025

Anthropic's new plugin system for Claude Code gives the agent new skills it can use to complete tasks.

SORA: CLIPPING TO CLONING

Sora enables everyone to become a potential creator of your content.

Subscribe now.

996 JUST MEANS YOU HAVE NO LEVERAGE

The reason why some people have to work 12 hours a day is because their ideas aren't good enough to survive on eight.

EVERYTHING IS TELEVISION

The capacity for solitude, sustained attention, and meaning that penetrates inward is being lost as people increasingly consider short-form video as the definition of American success.

WHY IT TOOK 4 YEARS TO GET A LOCK FILES SPECIFICATION

There have been over 1,800 posts on the subject of Python lock files since 2019.

Download the Enterprise AI Governance Kit from you.com

FIGURE AI'S NEW HUMANOID ROBOT CAN FOLD YOUR CLOTHES, DO THE DISHES

Figure 03 is a general-purpose humanoid robot that can perform a wide range of chores at home. It uses an AI-powered system called Helix to see and execute complex actions without a predetermined script. At 5 feet and 6 inches tall, the 132-pound robot can run for five hours on a single charge. The company aims to release the robot in 2026, but acknowledges that the goal is a big push. A video from Figure AI introducing the robot is available in the article.

INTEL DEBUTS NEW TECHNOLOGY IN MAKE-OR-BREAK MOMENT FOR CEO'S TURNAROUND BID

Intel's Panther Lake processor designs are in full production and will go on sale in laptops early next year. The chips are made with 18A technology, the most advanced chip production technology developed and employed in the US. Intel needs to show that new products will win back lost market share and attract customers to its foundry position. The Panther Lake processor will more readily balance a PC's need to run demanding software without rapidly draining batteries.

CHINA LAUNCHES WORLD'S FIRST DUAL-TOWER SOLAR-THERMAL POWER PLANT IN THE GOBI DESERT

China just launched a solar-thermal power plant that utilizes a dual-tower system to generate electricity in the Gobi Desert. The station features two 656-feet-high towers, each surrounded by a vast field of 27,000 mirrors. Known as heliostats, the mirrors concentrate sunlight onto the towers. The stored heat is used to generate steam, allowing the production of electricity well after sunset or during cloudy weather. The technology harnesses heat instead of sunlight, making it one of the few renewable technologies capable of providing stable, dispatchable energy on demand.

A BUILT-IN ‘OFF SWITCH' TO STOP PERSISTENT PAIN

EMBRACING THE PARALLEL CODING AGENT LIFESTYLE

Coding agent software is still really new. Models have only really become good enough to drive them in the past few months. One of the ways to use this software is to run multiple coding agents at once, sometimes in the same repo and against multiple checkouts or git work trees. This approach increases the number of tasks that can be fired off in parallel without adding too much cognitive overhead. This post discusses patterns for applying parallel agents effectively.

BITTER LESSONS BUILDING AI PRODUCTS

You shouldn't try to make AI work for your existing roadmap with fancy, clever engineering because much of it will be obsolete with the next major model upgrade. Instead, aim to understand model capabilities and how you can best pivot your roadmap accordingly. Companies need to learn to kill projects quicker if they need to hack together a solution to make up for intelligence. Just wait a few months and retry the idea when a new model drops.