Foundation models are shaping the future of computing. Model API spending has more than doubled in six months. Enterprises are increasing production inference rather than just model development, marking a shift from previous years. Code generation has become AI's first breakout use case. While open source continues to advance, Western labs have seen a slowdown in frontier breakthroughs, resulting in enterprise dollars consolidating around a few high-performing, closed-source models.
A walkthrough for building an AI assistant that can browse online clothing stores, locate specific garments, and generate virtual try-on previews showing how the clothes would look on you.
Buddhist scholars analyzed an AI-generated "sutra" created by ChatGPT o3, finding sophisticated philosophical concepts and imagery that effectively evokes core Buddhist teachings like emptiness through paradoxical deconstruction. However, they warned that while LLMs can recite dharma teachings, they lack true insight and may dangerously convince users they're receiving true divine guidance when prompted for spiritual advice.
Atlassian's new State of DevEx report [11]shows that while developers are saving time with AI, they're also facing more organizational complexity than ever. We surveyed developers and managers to uncover
The design philosophy for building AI applications is still in its infancy. One thing we can predict is that models will get much better. Designing AI applications to take advantage of this is probably the most important thing. It's important to understand your application structure, re-evaluate that structure as models improve, and make it easy to remove structure.
A Fractional Gaussian Filter and Pruning (FGFP) framework combines fractional-order differential calculus and Gaussian functions to compress neural networks for edge deployment.
Apple CEO Tim Cook's remarks on the company's third-quarter earnings call confirm that the company is going to invest more heavily in AI technology. He said that Apple is always looking to buy companies of any size that could help it develop its AI offerings. Apple is facing growing pressure from Wall Street to catch up to its Silicon Valley peers. In addition to growing its investment, Apple is also rearranging staff internally to focus more on AI.
Meta's shares are up 11% after earnings today, driven largely by the success of AI-powered ad recommendations, which drove 5% more conversions on Instagram and 3% on Facebook. Mark Zuckerberg told analysts the company will push aggressively on AI investments as the most aggressive AI timelines have held true.
Sachin Dev Duggal was ousted as CEO of $1.5 billion AI unicorn Builder.ai after investigators uncovered $249 million in fabricated sales and "round tripping" schemes. The company, which claimed AI could automate app development but relied heavily on human programmers in India, allegedly overstated revenue by 300% while Duggal transferred investor funds to personal accounts for luxury expenses.
Anthropic has released two major new features for Claude: the ability to draft and send emails, messages, and calendar invites directly from the Claude app, and the ability to upload PDFs, images, code files, and more to AI-powered apps.
GitHub's AI coding assistant hit 20 million cumulative users, adding 5 million since April and now serving 90% of the Fortune 100, with enterprise adoption up 75% quarter‑over‑quarter.
The SEC has unveiled “Project Crypto,” a directive to overhaul outdated securities rules and enable onchain trading of tokenized assets without unnecessary intermediaries. Chair Paul Atkins framed it as essential to making the US the “crypto capital of the world,” emphasizing regulatory updates that prevent driving innovators offshore and foster super-app ecosystems. Project Crypto builds on the President's Working Group report, with goals to create a cohesive framework that integrates blockchain technology into mainstream financial markets.
Coinbase reported $1.5 billion in Q2 revenue, up from $1.45 billion a year ago but below the $1.59 billion analysts expected, and saw adjusted EBITDA fall to $512 million from $596 million. Transaction revenue plunged 39% quarter-over-quarter to $764 million, despite Bitcoin and Ether hitting new yearly highs. The stock slid 7% in after-hours trading as waning trading volumes outpaced overall market strength. Coinbase highlighted progress in its institutional infrastructure, staking services, and Base network as strategic growth drivers beyond its core trading business.
Coinbase will roll out an onchain “everything exchange” in the coming months, enabling US users to trade tokenized real-world assets, including stocks, derivatives, prediction markets, and early-stage token sales. VP Max Branzburg says this one-stop shop aims to create a faster, more accessible, and truly global economy by bringing all asset classes onchain. The move positions Coinbase directly against Robinhood, Gemini, and Kraken, which already offer tokenized equities to non-US users.
Visa CEO Ryan McInerney announced multichain, multi-coin stablecoin settlement support on the Visa network, now spanning 4 blockchains and 4 stablecoins. New additions include EURC (euro-backed), USDG, and PYUSD via Paxos, along with support for Stellar and Avalanche chains. This infrastructure enables stablecoin settlement across 25+ fiat currencies globally, positioning Visa as a key layer for tokenized money movement across both banks and fintechs.
Digital Asset Treasury Companies (DATCOs) now hold over $100 billion in crypto – 791,000 BTC and 1.3 million ETH – led by Strategy (MSTR) and several emerging altcoin-focused entrants. They fuel their treasuries through At-the-Market equity programs and PIPE financings, creating an “accretive dilution” loop by raising capital at premiums to NAV and converting it into more digital assets per share. This model depends on persistently elevated equity premiums, and a collapse could trigger a rapid unwind, echoing the 1920s investment trust crashes and potentially pressuring asset prices. Looking ahead, DATCOs are set to diversify into new tokens, adopt hybrid operating models, and expand globally, remaining a potent structural source of buying pressure so long as market and capital-market conditions hold.
Ethereum's second decade is expected to transform the network from a smart contract platform into the foundational “trustware” layer powering global digital asset digitization, commerce, and identity. Institutions like GameSquare are accumulating ETH for its composability and community, while oracle networks and Layer 2s drive scalability and seamless real-world data integration. The protocol's evolution will make onchain interactions as invisible as AI-driven services, embedding Ethereum into enterprise and consumer applications.
Just as SOFR and Fed Funds rates anchor TradFi, DeFi now needs standardized benchmark rates - and Aave is becoming that foundation. With $58B in net deposits and over 50% of DeFi lending market share, Aave's rates reflect true market-wide credit conditions. Protocols like CoinDesk's Overnight Rates (CDOR), Cap, Level USD, and Plasma already use Aave's rates to price capital, route idle assets, or set hurdle yields. Its stability (e.g., only 20bps movement on a $100M deposit) enables floating-rate lending, structured products, and institutional risk management. Aave's dominance makes it DeFi's reference layer for rate discovery, liquidity, and scalability.
Zora has evolved from an NFT minting platform into a fully onchain social app, where every post and profile is a tradable ERC20. Creator coins pair with $ZORA and generate revenue on every trade (over 1.6M coins have been created across 200K+ creators, with total trading volume surpassing $445M). To power this mobile-native UX, Zora uses Privy to embed self-custodial wallets directly into the app, enabling users to onboard via content instead of crypto complexity. With over 4M wallets and 2M+ monthly actives, Zora's model blurs the line between social media and financial protocol, turning followers into participants and content into capital.
Moonwell's Flagship USDC Vault on Base is yielding a net 10.15% APY, with rates powered by Morpho (+3.05%), WELL (+1.10%), OP rewards (+1.07%), and a native base rate of 5.78%.
Figma has gone public on the New York Stock Exchange after a prolonged delay, with its stock price surging so rapidly that trading was briefly halted due to volatility. Shares initially priced at $33 quickly spiked to between $101 and $112, giving the company a market capitalization of $45 billion. The blockbuster IPO marks a major rebound from its failed $20 billion acquisition by Adobe in 2023, now seen as a minor blip in its growth story.
Google announced a new AI shopping feature that suggests outfits and home decor ideas while connecting users with similar products. The feature expands Vision Match technology from single garments to complete outfit and room design recommendations for U.S. users this fall. Google also launched its "try it on" tool nationwide, allowing users to upload photos and see how clothing items might look.
Uber Eats has introduced a suite of AI-powered features aimed at improving the customer experience, including AI-generated menu descriptions, enhanced food images, and summarized reviews. The platform can now digitally upgrade low-quality food photos and even generate missing visuals using AI, while also incentivizing users in select countries with in-app credits for uploading their own meal photos. Additionally, a new Live Order Chat allows restaurants to communicate directly with customers to resolve issues or clarify requests in real time.
Designers can maintain business curiosity while building recognizable positioning through alternative strategies beyond narrow specialization. Options include maintaining high demand with limited supply or adopting low-cost, high-volume models like major retailers. The preferred approach involves maximizing business profitability to pursue a variety of experiences outside professional work.
This guide outlines essential requirements for making business videos ADA-compliant to serve over 70 million people with disabilities in the US. Key accessibility features include accurate captions, transcripts, audio descriptions, keyboard navigation, and proper contrast ratios. Non-compliance risks legal settlements, mandatory remediation, and brand damage as courts use WCAG 2.2 standards.
Operating beyond your circle of competence often leads to second-guessing and uncertainty, while staying within it enables faster, more confident decisions and greater team trust. Regularly revisiting fundamentals, tracking outcomes, seeking feedback, and reflecting on strengths, gaps, and unknowns helps sharpen that circle and build long-term impact through steady, intentional growth.
AI transforms developer handoff workflows by automating repetitive tasks and bridging communication gaps between design and development teams. This automation reduces errors, saves significant time, and ensures better alignment throughout the design-to-code process. Teams implementing AI-powered handoff tools report saving up to 75 days of engineering time within six months.
Studio Glass developed a deeply rooted brand identity for the New Forest by combining historical research, local collaboration, and folklore to foster a sense of stewardship rather than simple tourism. Drawing from archival materials, seasonal aesthetics, and local voices, the brand strategy bridges centuries of heritage with modern storytelling, resulting in a unifying, community-driven visual and editorial experience.
System design is often taught through domain-specific solutions, making it difficult to identify shared principles that span across fields like databases, operating systems, and architecture. A unified, concise vocabulary of design principles would help clarify trade-offs, support cross-domain comparisons, and improve communication among students, researchers, and practitioners.
Modern lighting designs transcend mere illumination, combining artistic expression with functionality through innovative use of color, form, and materials.
Sofia Bonati creates bold, imagined portraits where women gaze confidently through geometric patterns and natural details like insect wings, blending nostalgic and surreal elements with contemporary style and vibrant backgrounds.
It's time to ditch the system silos and slow workarounds. See how API integration can help your business work smarter—not harder. From cutting out manual work to accessing real-time insights across your systems, this guide explains how APIs help build a backbone for lean, agile operations.
Following its massive IPO debut, Figma's value surged to nearly $68 billion, with initial investors like Index Ventures, Greylock, Kleiner Perkins, and Sequoia holding $24 billion in stock. The IPO raised $1.2 billion, with most proceeds going to existing investors, although a lock-up period restricts major sales until January. This marks a shift for the tech IPO market, previously slow due to inflation and market conditions, with Figma standing out as a significant success story.
AI infrastructure company fal raised $125 million in a Series C round, led by Meritech and joined by Salesforce Ventures, Shopify Ventures, and Google AI Futures fund, valuing it at $1.5 billion. Fal specializes in running multimodal models for enterprises, such as image generation AI for ecommerce and advertising. Enterprise clients use fal to create infinite ad iterations and different product images, providing significant value through diverse asset customization.
After 2021's ZIRP era, tech executives swore they'd never miss another liquidity window - yet here we are with barely any IPOs despite markets at all-time highs. Since the election, speculative stocks like ARKK have outperformed the Mag7 by 5x. Companies keep waiting for the "perfect" IPO window while multiples march back toward 2021 levels, and recent IPOs like Rubrik and Circle show strong performance. The few who prepared early are capitalizing - Figma's is massively oversubscribed. In an industry where barriers to competition are falling and public markets are begging for new names to bid on, waiting for the perfect moment is the exact mistake everyone promised they wouldn't repeat.
David Cummings shares the memory of securing office space with a foosball table early in his startup journey. He recounts signing up for a credit card to purchase a foosball table after relocating to a new office space due to cash constraints. The story highlights the evolving nature of startup culture and the importance of cherishing early entrepreneurial experiences.
Executing a JTBD (Jobs to be Done) interview process revealed that interview difficulty stems from the need for ethnographic research skills and overcoming human memory limitations. The recommendation is to focus on recent buyers and conduct partner interviews to improve accuracy. JTBD insights at Commoncog discovered two customer paths to membership, revealing a predominant conversion via paywalled articles within series, challenging prior assumptions and guiding strategic focus.
Everyone's panicking about how to measure ARR with AI's usage-based pricing, but ARR was never that clean. Snowflake is entirely usage-based yet valued as recurring revenue. Datadog mixes flat fees with variable metrics. Even traditional SaaS has messy pricing - customers can be charged by seats, active users, and usage tiers all at once. What makes revenue "recurring" isn't whether it's fixed monthly, it's whether you can predict it with confidence. Your job is to define ARR honestly for your business model, not game it to fit some perfect definition.