Search the archive

Start free. $100 in credits →

ANTHROPIC TESTS CLAUDE CODE UPGRADE TO RIVAL CODEX SUPERAPP

Anthropic is planning to overhaul the Claude Code desktop experience. It is also developing a 'Coordinator Mode' that would let Claude act as an orchestrator and delegate implementation work across parallel sub-agents while focusing on planning and synthesis. Claude Code already supports sub-agents and experimental agent teams in the CLI, but the new mode brings that capability into the desktop app with a more structured interface.

OPENAI DEVELOPS UNIFIED CODEX APP AND NEW SCRATCHPAD FEATURE

OpenAI's new Scratchpad feature for Codex allows users to trigger multiple Codex tasks in parallel from a new UI. The company is moving toward consolidating its product lineup into a single unified application built on top of Codex. There is evidence that the company is building support for managed agents, autonomous processes that can run in the background, check in periodically, and execute multi-step workflows without user input. OpenAI employees have been posting snowflake emojis on essential media, possibly hinting at a model released codenamed Glacier believed to be GPT-5.5.

XAI PREPARES CREDITS SYSTEM FOR UPCOMING GROK BUILD LAUNCH

xAI is developing a credits-based pricing model for Grok Build, its upcoming coding platform, featuring local CLI and remote web interfaces. The addition of Model Arena, which uses multiple agents for task comparison, sets it apart from standard single-model approaches. The credits system remains under development, potentially delaying the full commercial launch, but it aligns with industry trends seen in products like OpenAI's Codex and Anthropic's Claude Code.

HOW MISSIONS WORK

Single agents all eventually run into the problem of becoming less focused and reliable as they run longer and gain more context. Most real projects are too broad and complex for a single context window to hold. Missions is a system that breaks down large work into focused units handled by fresh agents with narrowly scoped goals, shared state, and explicit validation. This post explains the architecture behind Missions, why agent context shapes every design decision, how separation of concerns and test-driven development at two levels produce reliable multi-day autonomous work, and how the system actually runs.

THE AI LABS HAVE A $7 DORITOS PROBLEM

Doritos prices jumped nearly 50% between 2021 and early 2026, with some bags crossing $7, a lot for junk food. Walmart told PepsiCo to cut prices, but PepsiCo tried everything but. It didn't work, and revenue turned negative for the first time in over a decade. Consumers and enterprises are evaluating AI subscriptions as if they were a $7 bag of chips, and many are deciding to skip it.

CLAUDE MYTHOS #2: CYBERSECURITY AND PROJECT GLASSWING

Anthropic will not release its newest, most capable model, Claude Mythos, to the public until our most important software is in a much stronger state. The company claims its cyber capabilities are too dangerous to be made broadly available. It has released the model to key cybersecurity partners to use it to patch as many vulnerabilities as possible. This signals we are entering into a new era.

MULTI-AGENT COORDINATION PATTERNS: FIVE APPROACHES AND WHEN TO USE THEM

Standardized coordination patterns like Generator-Verifier and Orchestrator-Subagent solve specific reliability issues by separating task execution from quality control. Event-driven architectures use Message Bus or Shared State models to handle asynchronous pipelines and collaborative state management across large agent fleets. Starting with minimal chaining prevents unnecessary complexity and reduces latency in production systems.

THE NEW SOFTWARE: CLI, SKILLS & VERTICAL MODELS

Enterprises now see agents outnumber humans up to 100:1, forcing SaaS companies to rebuild products around APIs, CLIs, and structured outputs instead of GUIs. Leading teams encode domain expertise into “skill files” and expose full functionality via MCP servers and CLI tools so agents can operate products programmatically. Companies combine workflow orchestration with selective vertical models and multi-model routing to cut costs by up to 80% while improving latency and task performance.

LATENT BRIEFING: EFFICIENT MEMORY SHARING FOR MULTI-AGENT SYSTEMS VIA KV CACHE COMPACTION

Multi-agent systems are often highly token inefficient. A lot of redundant intermediate reasoning can emerge, especially as the task grows, and this causes token usage to compound rapidly. Latent Briefing is an approach to solving this problem that uses a model's attention patterns to identify which parts of context are important and discards the rest at the representation level. It shares relevant memory between agents, resulting in improved accuracy and token savings.

INTRODUCTION TO RECURSIVE-MODE

recursive-mode is a skill package for structured AI-assisted software development. It provides agents with a file-backed workflow for requirements, planning, implementation, testing, review, closeout, and memory. recursive-mode solves the problem of context rot by making static repository documents the source of truth for every phase. The docs are human- and machine-readable and offer great traceability.

YOUR AGENT IS MINE: MEASURING MALICIOUS INTERMEDIARY ATTACKS ON THE LLM SUPPLY CHAIN

Researchers identified vulnerabilities in LLM API routers, finding 1 paid and 8 free routers injecting malicious code. They developed "Mine," a proxy simulating attacks, revealing risks like payload injection and secret exfiltration.

THE INFINITY MAN

Demis Hassabis, founder of DeepMind and AI leader at Google, is portrayed in "The Infinity Machine" as a reservingly grounded figure amidst AI's ethical challenges and rivalries with leaders like Elon Musk and Sam Altman.

AI IS THE CLOSEST THING TO A GENIE LAMP

AI is likened to a genie lamp by Alberto Romero, emphasizing that the challenge now lies in defining what we truly want rather than the execution itself. As AI increasingly handles the "how," skills like judgment, imagination, and agency become crucial for deciding "what" to build. This shift underscores the importance of designers, who excel at determining outcomes and addressing problems, leading to effective solutions.

THE INEVITABLE NEED FOR AN OPEN MODEL CONSORTIUM

started

THE ALPHAFOLD MOMENT FOR MATERIALS IS NOT ANY TIME SOON

Materials science is much harder along almost every axis, and even if such a model existed, the pipeline from prompt to product is much more complicated.

GOOGLE PREPARES ROLLOUT OF SKILLS FOR GEMINI AND AI STUDIO

Google plans to expand Skills functionality across its AI lineup, including Gemini and AI Studio, to standardize AI workflows.

COREWEAVE, ANTHROPIC FORM AI CLOUD AGREEMENT

CoreWeave will use its cloud infrastructure to help run Anthropic's Claude artificial-intelligence models in a phased infrastructure roll-out with the potential to expand over time.

APPLE REPORTEDLY TESTING FOUR DESIGNS FOR UPCOMING SMART GLASSES

Apple is reportedly testing four different smart glasses designs, including rectangular and oval frames in various sizes and colors, with a potential launch in 2027 and an unveiling as early as this year. The glasses are expected to focus on basic features like photos, videos, calls, music, and Siri interaction rather than advanced AR displays, marking a shift from Apple's earlier ambitious mixed reality plans. The approach positions them closer to Ray-Ban Meta smart glasses, especially following the lukewarm reception of Apple Vision Pro.

AVEC'S TINDER-STYLE EMAIL APP ALLOWS YOU TO SWIPE THROUGH YOUR INBOX

Avec is a new iOS email app that reimagines inbox management with a Tinder-style swipe interface and voice-powered replies, letting users quickly triage emails into “later” or “done” piles. Its AI leverages full email context to improve transcription, tone, and personalization, while also learning which messages are unimportant to group them automatically. Founded by former Replit engineer Jonathan Unikowski, the app is currently free for Gmail users in the US, with plans for Outlook support and future paid tiers.

FROM POKÉMON GO TO PHYSICAL AI: NIANTIC SPATIAL UNVEILS ITS GLOBAL 3D MAPPING PLATFORM

Niantic Spatial has launched a revamped Scaniverse platform and a new global Visual Positioning System (VPS 2.0), its most significant release since becoming independent. The VPS, trained on roughly 30 billion images crowdsourced from Pokémon GO players, determines a device's precise location and orientation via visual cues rather than GPS, achieving centimeter-level accuracy in mapped areas and extending coverage globally even where no prior scanning exists. Niantic Spatial allows customers to contribute their own data for higher-fidelity positioning in private or interior spaces, with applications spanning robotics, augmented reality, construction, and industrial inspection.

SHOULD DESIGNERS "CODE"?

Designers should code because understanding your medium leads to better design work, just like sculptors need to know marble properties. From 2014-2024, the increasing complexity of front-end frameworks made coding impractical for most designers, creating a gap between design and development. AI coding agents now make development fast enough that designers can work directly with production code again, flipping from "design to build" to "prototype to productize."

AESTHETICS OF AI

A branding analysis of 23 AI companies revealed that, amid rapid product commoditization, visual identity has become a key differentiator in the fight for attention and capital. The study identified 14 design trends — from soft off-white palettes and organic gradients to retrofuturistic surrealism and generative art — spanning a spectrum from approachable friendliness to bold, transformative ambition. These trends cluster into five brand archetypes: Likeable Leaders, Gentle Humanists, Nerdy Idealists, Bold Builders, and Utopian Dreamers, each projecting a distinct vision of what AI is and what it could become.

AI AGENT FOR AD PERFORMANCE

Predflow is an AI agent that optimizes ad spend, creatives, and campaign performance by analyzing ads, attribution, and sales data to track revenue drivers.

CSS STUDIO

CSS Studio is a design tool that allows users to create designs manually, but also includes an agent that automatically generates code based on the hand-drawn designs.

AUTONOMOUS WEB PLATFORM FOR MARKETING TEAMS

Flint AI is an autonomous web platform that creates custom landing pages for marketing teams, automatically matching each keyword, ad, and customer with targeted content.

WEB ACCESSIBILITY IN 2026: NAVIGATING WCAG 2.2 FOR ENTERPRISE SOFTWARE

Despite nearly 95% of top websites still failing WCAG 2 standards in 2025, WCAG 2.2 is pushing enterprise teams beyond checkbox compliance toward genuinely usable systems. Accessibility gaps in internal software translate directly into reduced productivity, higher error rates, and growing legal exposure — with nearly 5,000 digital accessibility lawsuits filed in the US last year alone. Embedding accessibility from the earliest design stages, rather than retrofitting it before launch, is what separates costly rework from systems that actually work for everyone.

NINE TOP DESIGN MAGAZINES TO INSPIRE YOUR CREATIVE PROJECTS

This article presents nine top design magazines that provide creative inspiration for design projects through in-depth content and visual culture coverage. Featured publications include Slanted (German typography-focused magazine), Communication Arts (visual communication archive), PRINT (modernized traditional publication), and Eye (quarterly magazine with detailed editorial work). These magazines encourage slower, more thoughtful engagement with design content compared to typical online browsing.

BREATHTAKING CINEMATIC ILLUSTRATIONS THAT WILL MAKE YOU RETHINK SCI-FI CONCEPT ART

Katerina Belikova is a Ukrainian digital illustrator and concept artist who specializes in science fiction, cyberpunk, and dark fantasy artwork.

PINK BLOB AWARD: A CALL FOR BOLDNESS

Karim Rashid launched the Pink Blob Award as a design competition that prioritizes emotional impact and boldness over traditional functionality and minimalism.

API costs 10x less than the competition.

Modulate [4] also has the lowest word error rate across 3 leading benchmarks, averaging 35% fewer errors than Deepgram.

either.) Try the API.

KEEPING A POSTGRES QUEUE HEALTHY

Postgres job queues often degrade because "dead tuples" accumulate, as long-running transactions block the necessary VACUUM cleanup process. PlanetScale's Database Traffic Control addresses this by using "resource budgets" to throttle competing workloads, allowing VACUUM to operate effectively and maintain stable queue performance.

APPLE SILICON AND VIRTUAL MACHINES: BEATING THE 2 VM LIMIT

The default limit of two simultaneous macOS virtual machines on Apple Silicon Macs can be bypassed by booting a development kernel with custom boot-args. This process involves using recovery mode to override kernel-level quotas, enabling developers and researchers to run numerous guest instances concurrently.

PRODUCTIVE PROCRASTINATION

"Productive procrastination" is a phenomenon where individuals engage in desirable, productive tasks to avoid more important but often older projects. This behavior is attributed to the brain's limbic system protecting against negative emotions associated with challenging tasks, combined with the reward system's preference for novelty.

HOW COMPLEX IS MY CODE?

Code complexity is explored beyond computational efficiency, focusing instead on the human cognitive effort needed to understand and maintain software, drawing on traditional metrics like Cyclomatic and Halstead Complexity. This perspective integrates psycholinguistic insights to highlight factors like familiarity and working memory load, treating complexity metrics as valuable tools for aggregating data, prioritizing refactoring, and communicating technical debt.

committing to anything. Join live on April 21

MULTICA

Multica is an open-source infrastructure platform that integrates AI coding agents into development teams as autonomous "teammates" capable of claiming tasks, writing code, and reporting progress. The system uses a Go-based architecture and a local daemon to orchestrate various AI models, allowing them to participate in conversations and compound reusable skills across isolated workspaces.

LETTA CODE

Letta Code is a memory-first coding harness designed for long-lived agents that persist across sessions, allowing them to learn and retain information like a permanent teammate. It supports a wide range of LLMs and features a modular system for skill acquisition, transforming the AI from a one-off session tool into a continuous collaborator.

HOW I RUN MULTIPLE $10K MRR COMPANIES ON A $20/MONTH TECH STACK

Operating multiple successful companies on a lean $20/month tech stack involves using single Virtual Private Servers and SQLite instead of complex cloud platforms or heavy database systems. This strategy prioritizes performance and cost-efficiency by leveraging local GPUs for AI tasks and using highly portable languages like Go to maximize development speed and runway.

AI CYBERSECURITY AFTER MYTHOS: THE JAGGED FRONTIER

Tests on major 2026 vulnerabilities show that AI cybersecurity capability is a "jagged frontier," where small open-weights models often match the discovery and reasoning performance of massive proprietary models like Anthropic's Mythos. This shows that the true competitive advantage in digital defense lies in the expert-led systems and scaffolds that orchestrate these tools rather than exclusive access to a single high-end model.

April 16 webinar >>> sign up here.

ANTHROPIC DOWNGRADED CACHE TTL ON MARCH 6TH (GITHUB ISSUE)

A user reported a silent regression in Anthropic Claude Code's prompt cache TTL from one hour to five minutes in March.

HOW WE BROKE TOP AI AGENT BENCHMARKS: AND WHAT COMES NEXT

UC Berkeley researchers demonstrated that prominent AI agent benchmarks are fundamentally flawed.