Search the archive

Subscribe now.

POSTGRESQL 18 EYES ANALYTICS BOOST AND DISTRIBUTED FUTURE

PostgreSQL 18, set to be released later this month, will introduce a new asynchronous I/O subsystem that will allow read statements to be called from disk more quickly.

KRUCI: POST-MORTEM OF A UI LIBRARY

kruci was a declarative, imperative user interface library inspired by Swift UI that was abandoned due to unsolvable constraints regarding performance, memory, and simple practicality.

IPHONE 17, THE ‘THINNEST IPHONE EVER,' AND EVERYTHING ELSE WE'RE EXPECTING OUT OF APPLE'S HARDWARE EVENT

Apple's new hardware event will start at 10 AM PT on September 9.

WHY THE BROWSER COMPANY IS BEING ACQUIRED TO BRING DIA TO THE MASSES

Atlassian will provide The Browser Company with the resources to keep building without compromising its values.

NEURALINK EXPANDS TO CANADA WITH SUCCESSFUL BRAIN IMPLANTS IN SPINAL INJURY PATIENTS

Two people in Toronto have received Neuralink brain implants.

AN INTERVIEW WITH CLOUDFLARE FOUNDER AND CEO MATTHEW PRINCE ABOUT INTERNET HISTORY AND PAY-PER-CRAWL

In this interview, Cloudflare Founder and CEO Matthew Prince discusses his background, the original Cloudflare idea, what Cloudflare is today, and the opportunistic way in which it became the company that it is.

REGISTER TODAY

OPENAI'S JOBS PLATFORM

OpenAI has introduced a new hiring platform powered by AI to match companies with AI-fluent talent and launched OpenAI Certifications to validate AI skills.

ATLASSIAN TO BUY ARC DEVELOPER THE BROWSER COMPANY FOR $610M

Atlassian is seeking to create an AI-powered browser optimized for knowledge workers, building off its existing suite of productivity SaaS applications. The Browser Company will continue developing its upcoming Dia browser under independent operations.

CHINA'S DEEPSEEK PREPS AI AGENT FOR END-2025 TO RIVAL OPENAI

The Chinese startup is developing a multi-step AI agent software similar to OpenAI's Agent tool that can handle complex tasks without human supervision.

AI ARTISTS VS. AI ENGINEERS

AI Artists give AI systems full creative control, while AI engineers operate under constraints to optimize for quality. The distinction between the two is a spectrum, and most people fall somewhere in the middle, closer to one side or the other. While large language models are very powerful fools, in the absence of the ability to do fine-grained learning in the way a human does, they need stricter guardrails than a truly general-purpose system would ideally want. Companies should focus on building trust with customers by doing well-scoped, expected things.

A PM'S GUIDE TO AI AGENT ARCHITECTURE: WHY CAPABILITY DOESN'T EQUAL ADOPTION

Users tend to abandon agents after their first real problems. The real challenge for product managers is making architectural decisions that shape how users experience and begin to trust the agent. This post walks through the different layers of AI agent architecture to help readers understand why some agents feel magical while others feel frustrating. Product managers should architect for the magical experience.

BYTEDANCE'S ROBOTIC REASONING

Robix is a unified vision-language model for robots that combines reasoning, task planning, and natural language interaction in one framework. It enables robots to handle complex, long-horizon tasks and communicate naturally with humans through verbal responses and proactive dialogue.

GOOGLE RELEASED EMBEDDINGGEMMA

Google's EmbeddingGemma is a compact, multilingual embedding model with 300M parameters optimized for on-device performance.

OPENAI TO LAUNCH ITS FIRST AI CHIP IN 2026 WITH BROADCOM

OpenAI and Broadcom have partnered to produce an artificial intelligence chip. The chip will be used internally at OpenAI and won't be made available to external customers. OpenAI's move follows efforts by Google, Amazon, and Meta, which have all built custom chips to handle AI workloads. Broadcom expects AI revenue growth for fiscal 2026 to improve significantly.

SIERRA RAISES $350M AT $10B VALUATION TO EXPAND AI AGENT PLATFORM

Sierra announced raising $350M at a $10B valuation to expand its AI agent platform, backed by Greenoaks. The platform supports hundreds of major companies, reaching 90% of Americans in retail and 50% of US families in healthcare. New funding will enhance AI development capabilities and international expansion, strengthening Sierra's global presence.

Subscribe now.

GOOGLE PHOTOS ADDS VEO 3 VIDEO GENERATOR

Google Photos has integrated its latest video generation model, Veo 3, into the new Create tab, enhancing its image-to-video capabilities.

GOOGLE'S JULES SHIPS CODE REVIEW UPGRADES WITH PR COMMENTS AND MORE

Jules can now reply to PR comments, create new repositories, and receive files.

NASDAQ TIGHTENS SCRUTINY OF CRYPTO TREASURY RAISES

Nasdaq is increasing oversight of U.S.-listed firms raising capital to buy crypto, now requiring shareholder votes and greater disclosures, with potential suspension or delisting for noncompliance. The move follows an increase in digital asset treasury (DAT) raises, with 154 firms announcing $98.4B in planned crypto purchases this year. The crackdown adds new hurdles for companies mirroring Strategy and BitMine's balance sheet playbooks, potentially slowing deal timelines.

'NFTS TURNED OUT TO BE A FAD,' SAYS KEVIN O'LEARY

TEMPO: A PAYMENTS CHAIN FROM STRIPE AND PARADIGM

Stripe and Paradigm have announced Tempo, a new payments-first blockchain specifically designed for stablecoins and real-world payments infrastructure that features major enterprise design partners, including OpenAI, Anthropic, Shopify, Visa, and Deutsche Bank. The blockchain offers 100,000+ TPS throughput with sub-second finality, EVM compatibility, and native support for payments and gas fees in any stablecoin through an enshrined AMM, along with payments-optimized UX features like dedicated payment lanes and memos. Tempo aims to bring traditional payment use cases onchain, including global payouts, payroll, remittances, and microtransactions, while maintaining principles of decentralization and stablecoin neutrality through an independent validator network.

FIREBLOCKS LAUNCHES STABLECOIN PAYMENTS NETWORK

Fireblocks has launched an enterprise-grade stablecoin network to help financial firms move USD-pegged tokens and build payment products, with Circle, Bridge, Zerohash, and Yellow Card among 40 participants. The network introduces unified APIs and workflows to streamline stablecoin transactions across providers, blockchains, and fiat rails, positioning Fireblocks as core infrastructure for global payments. With stablecoin markets projected to reach trillions, the firm is targeting significant institutional demand while expanding services in emerging markets like those in Africa.

WHY HYPE DATS CAN WIN

Digital Asset Treasury companies holding Hyperliquid's HYPE token possess a unique competitive advantage in the treasury company space through their ability to generate substantial yield via HIP-3, Hyperliquid's exchange infrastructure sharing mechanism. While BTC DATs like MicroStrategy rely primarily on brand and capital raising ability, and ETH/SOL DATs struggle with low staking yields of only 2-5%, HYPE DATs can potentially earn ~24% APY by staking tokens with exchanges built on Hyperliquid infrastructure and sharing in their trading fees. However, the strategy requires careful selection of exchanges to back since each will likely only accept around 1 million HYPE tokens, forcing HYPE DATs to make strategic bets on which exchanges will succeed in capturing significant trading volume.

SEPTEMBER YIELD FARMING PLAYBOOK

There are five high-yield DeFi strategies ranging from 12% to 42.5% APR, including single-sided Bitcoin exposure through GammaSwap Labs at 15% APY with minimal drawdown, and a complex HYPE token strategy using HyperEVM and Pendle that delivers 42.5% APR through leveraged looping. The strategies span various protocols and assets, from delta-neutral stablecoin vaults earning 12% APR to sophisticated USDe staking and looping strategies on Ethena Labs and Dolomite that can generate over 40% APR plus additional token rewards. Most strategies involve leverage, looping, or complex multi-protocol interactions to amplify returns, with associated risks including liquidation exposure and interest rate volatility that require active monitoring.

TRADING THE METAGAME: IN PURSUIT OF NARRATIVES

The cryptocurrency market has fundamentally shifted from valuing pure technology to prioritizing narrative and actual user adoption, with many venture-backed infrastructure projects failing because they lack compelling stories or real usage. Pump Fun's success revealed that most token buyers simply want price appreciation rather than technological innovation, exposing how previous cycles were sustained by speculation on unused blockchains rather than genuine utility. Market capitalizations are built on public trust and consumer psychology, requiring projects to either develop better tokenomics that resemble equity or eliminate tokens altogether. As crypto markets become more accessible to mainstream users, successful projects must understand that "normies" interpret value through viral narratives and social proof rather than technical specifications, making storytelling as crucial as the underlying technology.

MONETIZING ULTRA-NICHE KNOWLEDGE

Prediction markets represent the first system capable of making ultra-niche knowledge "liquid," allowing individuals with highly specific expertise to directly monetize information that previously had no clear economic value. Unlike traditional knowledge monetization methods that require building newsletters or convincing investors, prediction markets enable immediate trading on obscure insights, fundamentally changing the incentive structure around curiosity and specialized knowledge. This creates a dual benefit: it rewards individuals for pursuing seemingly useless interests while simultaneously making markets more efficient by incorporating diverse, granular information that was previously trapped in people's heads. The system transforms the "long tail" of human knowledge into valuable market signals, potentially creating the most efficient knowledge aggregation mechanism humans have developed.

PUBLIC COMPANIES NOW HOLD OVER 1M BITCOIN

Public firms collectively hold more than 1 million BTC, worth ~$111B, as Bitcoin treasury strategies accelerate.

CRYPTO MARKET POSITIONING

Crypto market positioning as of September shows a notable divergence between Bitcoin and Ethereum, with BTC spot positioning heavily long while ETH remains relatively neutral, though BTC's options put-call skew suggests underlying defensive sentiment.

ETHEREUM INTEROPERABILITY LAYER

The Ethereum Interop Layer (EIL) is a trustless cross-rollup interoperability solution designed specifically for Ethereum's L2 ecosystem that leverages their shared trust root to eliminate the need for trusted intermediaries.

COINBASE INCREASES LEVERAGE TO 50X ON INTERNATIONAL PERPS

Perpetual futures on Coinbase's international exchange have increased from a maximum of 20x leverage to 50x.

FIGMA SHARES SLIDE AS FIRST POST-IPO RESULTS MISS PROFIT EXPECTATIONS

Figma's shares dropped over 13% after-hours following the design software company's first quarterly earnings since its July IPO, where it broke even despite analyst expectations of 18 cents per share profit. The company reported $249.6 million in Q2 revenue (up 41% year-over-year) and launched four new products, including AI-powered prototyping tools and website publishing capabilities. Figma forecasts Q3 revenue of $263-265 million and full-year revenue of $1.021-1.025 billion, with strong customer metrics, including a 129% net dollar retention rate.

MORE PIXEL PHONES ARE GETTING GOOGLE'S NEW LOOK FOR ANDROID

Google's September Pixel drop expands the Material 3 Expressive design to Pixel 6 and newer devices, plus the Pixel Tablet, bringing a more colorful look and new personalization features. The update also adds Pixel Buds Pro 2 upgrades like Adaptive Audio, Pixel Watch auto-Maps display, AI-powered Gboard suggestions, Bluetooth audio sharing, a refreshed Quick Share, and the new Androidify app.

NEW AI MODEL TURNS PHOTOS INTO EXPLORABLE 3D WORLDS, WITH CAVEATS

Tencent's HunyuanWorld-Voyager is an open-weight AI model that transforms single images into explorable 3D-consistent video sequences by generating RGB video and depth information simultaneously. The system allows users to define camera movements through virtual scenes, producing 49-frame clips that maintain spatial consistency and can be chained together for more extended sequences. However, the model requires substantial computational power and faces geographic licensing restrictions, preventing use in the EU, UK, and South Korea.

HOW TO MAKE PEOPLE LISTEN TO YOU

When designers feel ignored, they should investigate root causes rather than blame dysfunctional organizations, identifying mismatches between their expectations and leadership goals. Common reasons include low organizational design maturity, business models that prioritize speed over quality, working on ideas leaders don't understand, and a lack of established trust with team members. The recommended approach involves conducting "listening tours" with colleagues, treating them like customers to build charm and find allies, then demonstrating value through small wins that create stories of problem-solving success.

APPROACHING DESIGN IN THE AGE OF AI

AI product design should prioritize user needs over technology, avoiding the common trap of defaulting to chat interfaces when AI's real value lies in inference and reasoning capabilities. Misinformation and hallucinations require design solutions, including user control over data sources, robust feedback systems, and consideration of smaller language models for specific contexts. The current moment presents both unprecedented opportunity and responsibility for implementing responsible AI principles like transparency and safety during this technological watershed.

ADAPT OR DIE? WHEN TOTAL BRAND OVERHAULS WORK, AND WHEN THEY DON'T

Heritage brand rebrands often spark backlash, as seen with Tate & Lyle's Golden Syrup and Cracker Barrel, where consumers resisted changes to long-familiar packaging. Experts suggest this sensitivity stems from people seeking stability amid broader societal flux, with nostalgia acting as an anchor. By contrast, brands like Mailchimp thrive on reinvention, balancing playful experimentation with consistency to connect with creative audiences. Jaguar's bold 2024 overhaul shows that rebrands can succeed not by pleasing everyone but by provoking strong reactions and signaling deeper philosophical shifts, cutting through a saturated design landscape.

CREATE AND EDIT DIAGRAMS WITH AI IN SECONDS

Collaborate with AI to create amazing diagrams. Draw flowcharts, architecture diagrams, hardware schematics, mind maps, etc, in seconds.

STUNNING PRESENTATIONS AT THE SPEED OF THOUGHT

Ghost is the AI-native slide editor that helps you go from idea to deck in minutes.

SCREEN RECORDINGS REIMAGINED WITH MOTION

Elevate your videos by creating professional, modern animated recordings. Capture every frame with Motion.

WHAT DESIGNERS CAN LEARN FROM THE FIRST INTERFACE — THE HUMAN FACE

The human face serves as the original interface, offering immediate, cross-cultural communication through micro-expressions that evolved as survival mechanisms. Digital interfaces attempt to replicate facial communication through emojis, avatars, and video calls, but lose the authenticity and spontaneity that make faces so powerful. Designers can learn from facial principles like timing, subtlety, and trust-building rhythms to create more natural digital experiences that feel intuitive rather than requiring constant explanation.

CHAT IS: THE FUTURE OR A TERRIBLE UI

AI-powered chat interfaces divide opinions between those who see them as software's future and others who view them as terrible UI. Proponents argue that chat leverages existing user familiarity, captures intent through natural language, and eliminates interface learning curves. Critics counter that chat lacks clear affordances, poorly displays complex information, and creates cumbersome scrolling experiences that lose context over time.

THE DESIGN OF SHALLOW THINKING

Our capacity for deep thought has been eroded less by AI and more by how the internet's design has evolved. Early online spaces felt like destinations that encouraged exploration, long-form expression, and reflection, while smartphones, infinite feeds, and algorithm-driven platforms shifted us toward shallow, compulsive consumption. These patterns were shaped by business models built on maximizing attention, not by accident. Rethinking design to include boundaries, stopping points, and depth-focused features could help restore sustained attention and more meaningful engagement.

3D ARTIST GAL YOSEF REIMAGINS POP CULTURE LEGENDS IN UNIQUE BLEND OF PLAYFULNESS AND PROVOCATION

Israeli 3D artist Gal Yosef transforms beloved cartoon characters into provocative digital sculptures that blend childhood nostalgia with sharp social commentary.

THE NEW CONJURING: LAST RITES POSTER IS THE PERFECT HOMAGE TO A HORROR CLASSIC

The new Conjuring: Last Rites poster pays direct homage to the iconic Exorcist design, featuring Ed and Lorraine Warren half-lit under a streetlamp against a dark, grainy backdrop.

JACOB HUTCHINSON MERGES THE STRUCTURE OF SWISS TYPE WITH THE SOFT IMPERFECTIONS OF OLD STAMPS AND VINYL STICKERS

Sheffield-born designer Jacob Huchinson merges tactile vintage print aesthetics with digital media, creating structured yet nostalgic visual identities, album art, and motion graphics for musicians and brands.