Search the archive

TRANSLATING

Product leaders should translate product metrics into clear financial impact when speaking with executives.

FINDING THE ROLE OF HUMANS IN AI PRODUCTS

As AI agents become increasingly autonomous, the human role shifts from doing tasks to supervising systems that do them.

APPLE'S FOLDABLE IPHONE TO FEATURE IPAD-LIKE INTERFACE WHEN OPENED

Apple's upcoming foldable iPhone will have an iPad-like layout when opened. It will be able to run apps side-by-side for the first time. The interior foldable display will be roughly the size of an iPad mini. The external screen will be about the size of the display on a small iPhone. Apple is developing new iOS app layouts so developers can adapt their software to the new interface.

MUSK UNVEILS JOINT TESLA-XAI PROJECT ‘MACROHARD,' EYES SOFTWARE DISRUPTION

Elon Musk has revealed a new joint project between Tesla and xAI called 'Macrohard' or 'Digital Optimus' aimed at creating a system capable of emulating the functions of software companies. The project pairs xAI's Grok with a Tesla-developed AI agent that processes real-time computer screen video and keyboard and mouse actions. The system will run on Tesla's in-house AI4 chip paired with xAI's Nvidia-based server hardware. xAI applied for a trademark for 'Macrohard' in August 2025.

ANDURIL, THE AUTONOMOUS WEAPONS MAKER, DOUBLES THE SIZE OF ITS SPACE UNIT

Anduril is acquiring ExoAnalytic Solutions, a space intelligence firm that operates a network of sensors that monitor the movements of satellites above Earth. The acquisition will more than double Anduril's staff dedicated to space projects. Anduril is best known for developing autonomous weapons such as cruise missiles, jet-powered missile interceptors, and underwater drones. The purchase of ExoAnalytic will accelerate Anduril's work on supporting national security space customers.

HOW TO DESIGN ANTIBODIES

AI-based tools can now help scientists design original antibodies on computers. Antibodies are among the most versatile tools in biology. Before AI, scientists searching for a useful antibody would have to screen billions of candidates in laboratory assays to identify just a handful with a high affinity for a target. Now, suitable binders can be found after just tens of attempts rather than billions. This post shows readers the modern computational process for designing antibodies from scratch.

how TLDR compares.

WORDPRESS EVERYWHERE

Thanks to advances in WebAssembly, it is now possible to spin up a web server, a database, and a full WordPress installation within a browser in about 30 seconds. WordPress has now launched a site that does exactly that. Users can launch it to build sites and cross-publish apps to the web, desktop, and iOS. WordPress plans to add features like peer-to-peer sync, version control integration, and cloud publishing.

TEMPORAL: THE 9-YEAR JOURNEY TO FIX TIME IN JAVASCRIPT

The JavaScript Temporal object, a full replacement for the Date object, enables date and time management in various scenarios, such as time zone and calendar representation, wall-clock time conversions, arithmetics, and formatting. It has an intricate and powerful API that exposes over 200 utility methods via several classes. Implementing Temporal was a challenge as it was a very large proposal that brought more challenges to JavaScript than any other proposal in the programming language's history. This post takes readers through the journey of how Temporal came to be.

PRODUCTIVE INDIVIDUALS DON'T MAKE PRODUCTIVE FIRMS

Individual AI is the vector by which most of the world's businesses will first experience the transformative magic of AI. However, productive individuals do not make productive firms. There is a gaping need for institutional intelligence that creates coordination, finds signal, creates objectivity, optimizes for edge, and scales revenue. Organizations need an institutional AI purpose-built for domain-specific problems that individual AIs can leverage as a tool.

THE MARGINAL HIRE

Companies are hiring again, but at a lower rate than before. AI is helping teams ship with smaller headcounts. Most organizations aren't laying workers off or restructuring, just hiring less than planned. AI is taking jobs, just invisibly, one unposted job at a time.

NVIDIA IS REPORTEDLY PLANNING ITS OWN OPEN SOURCE OPENCLAW COMPETITOR

Nvidia has been pitching NemoClaw to various corporate partners ahead of its annual developer conference next week.

AI MODELS ARE THE NEW REBAR

For the vast majority of inference calls, the model is interchangeable, and customers buy on price.

META PREPARING TO DEPLOY FOUR NEW HOMEGROWN CHIPS TO HANDLE AI

Meta's new chips are part of an effort to diversify its hardware sources, reduce reliance on outside chipmakers, and bring down costs.

AGAINST VIBES: WHEN IS A GENERATIVE MODEL USEFUL

Generative models are useful when the relative cost of encoding the work in a prompt and verifying the output is low, and the process used to complete the work doesn't matter.

WHEN THE CHAIN BECOMES THE PRODUCT

Token economics break product feedback loops by introducing a competing optimization target, leading teams to stop asking what developers need and focus on supporting the token narrative.

DEVELOPER TO FLEET COMMANDER

Developers take pride in the code they write, but builders take pride in what gets built, even if it was built through commanding a fleet of agents.

Read key findings and strategic insights from the retreat

REPLIT AGENT 4

Replit launched Agent 4, introducing an infinite design canvas and the ability to run parallel AI agents to build backends and frontends simultaneously. The update shifts from pure code generation to a collaborative product suite capable of shipping web apps, databases, and slide decks within a single integrated environment.

AI CODING STARTUP CURSOR IN TALKS FOR ABOUT $50 BILLION VALUATION

Cursor is in talks with investors for a funding round that would value it at $50 billion, nearly double its valuation last fall. The discussions are preliminary and may not result in a funding deal. One of the fastest-growing startups of all time, Cursor is a central player in tech's vibe coding era. It has attracted financing from top names in venture capital and the tech industry. The company's last financing round, announced in November, boosted its valuation to $29.3 billion.

PERPLEXITY'S PERSONAL COMPUTER LETS AI AGENTS ACCESS YOUR MAC MINI'S FILES

Personal Computer is a version of Perplexity Computer that handles things on a nearby computer. It handles requests from users and directs multiple AIs to complete tasks. It is effectively a project manager that delegates tasks to other AIs before combining the results and presenting them. Perplexity has specifically named the Mac Mini as a platform that Personal Computer can run on. There is currently a waitlist for access.

META PREPARING TO DEPLOY FOUR NEW HOMEGROWN CHIPS TO HANDLE AI

Meta has announced four new chips as part of an effort to diversify its hardware sources, reduce reliance on outside chipmakers, and bring down costs. The MTIA 300 is already in production for content ranking and recommendations training. The MTIA 400 has completed lab testing and is moving toward deployment. The MTIA 450 and MTIA 500 chips are scheduled for mass deployment in 2027.

DESIGNING AGENTS TO RESIST PROMPT INJECTION

OpenAI outlined how prompt injection attacks against agents increasingly resembled social engineering, and argued that defenses should focus not just on detecting malicious inputs but on limiting the impact of successful manipulation.

HOW A REGULAR PERSON CAN UTILIZE AI AGENTS

Non-technical users can build functional agents by deploying API connectors and project-based instructions to automate high-cognitive tasks. This modular framework utilizes parallel subagents to execute complex workflows, shifting the user role from conversational prompter to systems architect.

HOW AUTORESEARCH WILL CHANGE SMALL LANGUAGE MODELS ADOPTION

Autoresearch is a tool that ingests a training script and a metric and then edits code, runs experiments, checks if the metric improved, keeps or discards, and repeats. It allows researchers to optimize models overnight, even without any machine learning experience. Andrej Karpathy used it to squeeze 11% more speed out of his GPT-2 training. Shopify's CEO used the tool to train a 0.8B model overnight that outscored his previous 1.6B model.

NVIDIA'S NEMOTRON 3 SUPER

NVIDIA just released Nemotron 3 Super, a 120B open hybrid MoE model (12B active) designed for agentic and multi-agent AI systems.

REASONING EXPANDS FACTUAL RECALL IN LANGUAGE MODELS

Google's controlled experiments showed that enabling reasoning improved single-hop factual recall in LLMs by acting as a computational buffer and by generating related facts that help retrieve answers, though hallucinated intermediate facts increased final-answer errors.

THE DEPARTMENT OF WAR IS MAKING A HUGE MISTAKE

Large language models are probably not currently being used in mission-critical ways, but this will change within 20 years, when 99% of the workforce will be AIs. Our future civilization will run on AI labor. We need to think through some extremely important questions about who this future workforce will be accountable and aligned to, and who gets to determine that. The only way to preserve a free society is to make laws and norms saying it is unacceptable for the government to use AI to enforce mass surveillance, censorship, and control. Individual acts of corporate courage will not solve the problem.

INTRODUCING THE ANTHROPIC INSTITUTE

Anthropic has launched the Anthropic Institute to tackle AI's societal challenges. Guided by co-founder Jack Clark and an interdisciplinary team, key hires include Matt Botvinick for AI law, Anton Korinek for economic impacts, and Zoë Hitzig for economic connections. Anthropic has also expanded its Public Policy team to shape global AI governance, led by Sarah Heck with a new office in DC.

agents. Save with code TLDR75

DON'T CALL IT A MOAT

AI code generation relieves the execution bottleneck that previously functioned as a temporary barrier to entry for software companies.

CURSOR EXPANDS MARKETPLACE WITH 30+ NEW PLUGINS

Cursor added more than 30 marketplace plugins from partners, enabling the editor to read, write, and take actions across a wider range of developer tools and services.

NVIDIA TO INVEST $2 BILLION IN NEBIUS TO EXPAND AI CLOUD INFRASTRUCTURE

Nvidia and Nebius will work together to design and deploy large AI data centers.

ANTHROPIC GIVES CLAUDE SHARED CONTEXT ACROSS MICROSOFT EXCEL AND POWERPOINT, ENABLING REUSABLE WORKFLOWS IN MULTIPLE APPLICATIONS

The strategic move is aimed at expanding Anthropic's enterprise footprint and potentially challenging Microsoft's newly launched Copilot Cowork.

Try Awaken for free, and if you like it get $50 off

BITCOIN DROPS ON IRAN WAR UNCERTAINTY

Bitcoin pulled back to $69,500 during European trading hours after failing to penetrate $71,750 resistance, declining 0.55% since geopolitical tensions from the US-Israel war with Iran drove cautious trading and over $220 million in crypto futures liquidations (mostly longs), while most altcoins lagged. Derivatives positioning showed institutional appetite remaining weak with CME bitcoin futures open interest dropping to $7.39 billion (lowest since September 2024) and similar declines in ETH futures, though crypto sentiment improved from "extreme fear" to "fear" territory at 25/100 on the Fear and Greed index as the market demonstrated relative strength versus precious metals and US equities since the conflict began March 1.

WELLS FARGO FILES 'WFUSD' TRADEMARK FOR CRYPTO, PAYMENTS, AND TOKENIZATION

Wells Fargo, a financial services giant with approximately $2.1 trillion in assets, filed a trademark application for “WFUSD” on March 9 covering three international classes, including downloadable software for digital asset trading, payments and wallet functionality, cryptocurrency trading and exchange services, and software-as-a-service for asset tokenization and blockchain-based trading infrastructure. The “WFUSD” name resembles ticker symbols commonly used for stablecoins, and the filing explicitly references software for processing stablecoin transactions, though Wells Fargo has not publicly commented on the application. The trademark filing builds on Wells Fargo's prior crypto infrastructure investments, including $5 million in blockchain analytics firm Elliptic in February 2020 and participation in Talos's $105 million Series B round in May 2022 alongside Citigroup and BNY Mellon, which valued the crypto trading infrastructure firm at $1.25 billion.

AON TESTS STABLECOIN PAYMENTS FOR INSURANCE PREMIUMS

Aon, which manages roughly $5 trillion in assets, completed a proof-of-concept settling insurance premiums via USDC on Ethereum and PYUSD on Solana with Coinbase and Paxos – the first known instance of a major insurance broker accepting stablecoins for premium settlement. Cross-border premium settlement through bank clearing can take days, whereas stablecoin rails compress that to minutes with on-chain auditability. The initiative gained regulatory footing from the GENIUS Act, increasingly cited as a prerequisite for institutional stablecoin adoption at scale.

AI TOKENS RALLY AFTER NVIDIA OPEN-SOURCE AGENT PLAN

AI-linked crypto tokens rallied ~4.8% to a collective $14.17B market cap after reports that Nvidia is developing NemoClaw, an open-source platform for deploying autonomous, multi-step AI agents in enterprise environments. Bittensor (TAO) led gains, with NEAR Protocol and Internet Computer also advancing. Nvidia approached Salesforce, Cisco, Google, Adobe, and CrowdStrike as potential partners, with the platform incorporating enterprise security and privacy tooling. NemoClaw is slated to debut at Nvidia's GTC developer conference beginning March 17. Though there is no direct link with Nvidia's announcement and AI tokens, investors are hoping that open agents naturally use crypto rails for verifiability and decentralization.

SIX MISTAKES IN ERC-4337 SMART ACCOUNTS

This post identifies six critical ERC-4337 vulnerabilities, including improper access control, unsigned gas fields, and stateful validation. These flaws enable fund drainage or account hijacking. Developers must implement EIP-712 domain separation and strict EntryPoint restrictions to secure programmable smart wallets against sophisticated on-chain exploits.

CL-USDC ON BOROS: WHY WEEKEND FUNDING IS THE TRADE

CL-USDC perpetual funding data reveals a pronounced "TradFi Gap" phenomenon: weekends comprise only 29% of total hours yet account for 54% of all funding paid, with weekend mean APR of 40.82% running 2.85x above the weekday rate of 14.33% (overall mean 22.01%, positive funding frequency 79.2%). When TradFi closes, crude oil price discovery moves entirely onchain, and without desks to balance hedges, funding rates spike – the March 7 weekend averaged 74% APR as oil breached $100 amid Iran strike headlines. The proposed Boros trade is to enter pay-fixed/receive-floating swaps ahead of each weekend to capture the rate spike, then rotate to receive-fixed/pay-variable at Monday open as TradFi liquidity returns and rates compress.

SEVEN YEARS INSIDE A TOKEN-FUNDED VENTURE

Blockstack's $15.5 million SEC-qualified token offering shifted focus from user-centric tools like Gaia to chain infrastructure. This transition replaced empirical feedback loops with narrative-driven roadmaps, prioritizing token price over developer needs. It illustrates how crypto incentives can perpetually delay product-market fit by favoring future speculation over current utility.

AAVE FEBRUARY 2026 REPORT

Despite a February market correction that drove TVL down 21.58% MoM to $44.94B and triggered approximately $429M in liquidations across 12,500 transactions, Aave's protocol revenue paradoxically rose 31.24% MoM to $13.40M, with liquidation activity and Chainlink SVR contributing $4.6M (34.4% of monthly revenue). GHO demonstrated resilience, growing 6.74% MoM to $467.53M market cap, while RWA deposits crossed $1B and cumulative loans originated since V1 surpassed $1 trillion. Aave maintained 59.36% lending market share with MAU rising 34.88% MoM to 154.63K, and L2 chains accounting for 60%+ of active users. A governance proposal dubbed the "Aave Will Win Framework" was put forward to redirect 100% of product revenue to the DAO treasury.