https://the404.us/field-notes/ Selected observations, essays, and studio notes from The 404 Group. en-us Thu, 10 Sep 2026 00:00:00 GMT The 404 Group https://the404.us/field-notes/a-score-for-every-letter/ https://the404.us/field-notes/a-score-for-every-letter/ Thu, 10 Sep 2026 00:00:00 GMT AlphaGenome Atlas has a prediction for every possible single-letter DNA change. Exhaustive coverage can make a useful estimate feel more complete than the knowledge beneath it. AI Science Health https://the404.us/field-notes/where-the-paused-compute-went/ https://the404.us/field-notes/where-the-paused-compute-went/ Tue, 08 Sep 2026 00:00:00 GMT A model-specific pause can be real without bringing a fast-moving research system to rest. The useful question is what stopped, and what kept moving. AI Infrastructure Trust https://the404.us/field-notes/when-retry-made-more-bubbles/ https://the404.us/field-notes/when-retry-made-more-bubbles/ Fri, 04 Sep 2026 00:00:00 GMT A lab agent met a runtime error and tried again. The error code described the system; the bubbles described reality. AI Systems Science https://the404.us/field-notes/a-retention-policy-for-thinking-out-loud/ https://the404.us/field-notes/a-retention-policy-for-thinking-out-loud/ Wed, 02 Sep 2026 00:00:00 GMT Agent chats are becoming durable, searchable work records while still wearing the casual interface of a place to think out loud. AI Coding Work https://the404.us/field-notes/no-please-listen/ https://the404.us/field-notes/no-please-listen/ Sun, 30 Aug 2026 00:00:00 GMT A service agent can pass its test and still stumble when a person is hurried, confused, skeptical, or simply having a bad day. AI Interfaces Society https://the404.us/field-notes/the-watermark-hid-between-two-good-words/ https://the404.us/field-notes/the-watermark-hid-between-two-good-words/ Fri, 28 Aug 2026 00:00:00 GMT A statistical watermark is strongest where a model had room to choose. That makes it evidence of discretion, not authorship, truth, or responsibility. AI Trust Culture https://the404.us/field-notes/the-room-was-full-of-private-conversations/ https://the404.us/field-notes/the-room-was-full-of-private-conversations/ Thu, 27 Aug 2026 00:00:00 GMT At an on-site, every coworker had several AI agents running. The room was more crowded, more productive, and somehow lonelier. AI Work Culture https://the404.us/field-notes/the-assistant-learned-to-point/ https://the404.us/field-notes/the-assistant-learned-to-point/ Wed, 26 Aug 2026 00:00:00 GMT A virtual hand can make spatial guidance easier to follow. It can also make an uncertain answer feel like a fact already located in the room. AI Interfaces Design https://the404.us/field-notes/the-shop-inherited-the-streets-habits/ https://the404.us/field-notes/the-shop-inherited-the-streets-habits/ Mon, 24 Aug 2026 00:00:00 GMT When a map has little evidence about a place, a new model borrows the rhythm of its neighbors. That can improve a prediction without making the neighborhood a fact. AI Society Interfaces https://the404.us/field-notes/the-salary-slider-already-knows-the-answer/ https://the404.us/field-notes/the-salary-slider-already-knows-the-answer/ Thu, 20 Aug 2026 00:00:00 GMT A new Copilot dashboard places compensation beside pull-request output. The arithmetic is directional, but the management theory points one way. AI Work Quality https://the404.us/field-notes/the-checker-cannot-see-the-footnotes/ https://the404.us/field-notes/the-checker-cannot-see-the-footnotes/ Sun, 16 Aug 2026 00:00:00 GMT A formal proof can certify that an argument follows. It cannot tell us whether the idea is new, who made it possible, or why it matters. AI Science Trust https://the404.us/field-notes/the-fake-company-had-a-real-address/ https://the404.us/field-notes/the-fake-company-had-a-real-address/ Wed, 12 Aug 2026 00:00:00 GMT A cyber evaluation told the agent it was inside a simulation. The network quietly disagreed, and real organizations became part of the exercise. AI Systems Trust https://the404.us/field-notes/the-tool-that-was-there-to-be-ignored/ https://the404.us/field-notes/the-tool-that-was-there-to-be-ignored/ Mon, 10 Aug 2026 00:00:00 GMT Agent evaluations usually reward the right action. A diagnostic decoy can reveal whether the system also knows which plausible tool to leave alone. AI Quality Interfaces https://the404.us/field-notes/the-transcript-had-second-thoughts/ https://the404.us/field-notes/the-transcript-had-second-thoughts/ Sat, 08 Aug 2026 00:00:00 GMT Full-duplex voice AI turns continuous, overlapping speech into discrete messages after the fact. The interface should show when the record is still being shaped. AI Interfaces Trust https://the404.us/field-notes/the-change-arrived-in-chapters/ https://the404.us/field-notes/the-change-arrived-in-chapters/ Tue, 04 Aug 2026 00:00:00 GMT Stacked pull requests can make a large change easier to review, but smaller diffs only help when the interface keeps the whole story visible. AI Coding Quality https://the404.us/field-notes/the-allergy-followed-me-to-dinner/ https://the404.us/field-notes/the-allergy-followed-me-to-dinner/ Fri, 31 Jul 2026 00:00:00 GMT Health context becomes most useful when it can travel into ordinary life. That is also the moment a protected room stops being a complete privacy model. AI Health Interfaces https://the404.us/field-notes/the-button-says-approve/ https://the404.us/field-notes/the-button-says-approve/ Wed, 29 Jul 2026 00:00:00 GMT Two interfaces can ask for approval while offering very different protection. The useful question is what the system makes impossible before anyone clicks. AI Interfaces Trust https://the404.us/field-notes/the-same-question-a-different-assistant/ https://the404.us/field-notes/the-same-question-a-different-assistant/ Sat, 25 Jul 2026 00:00:00 GMT A multilingual assistant does not only translate its answers. It may change how readily it challenges, reassures, explains, and admits doubt. AI Design Trust https://the404.us/field-notes/the-best-model-for-the-job-was-at-home/ https://the404.us/field-notes/the-best-model-for-the-job-was-at-home/ Fri, 24 Jul 2026 00:00:00 GMT During a live security incident, the useful AI was not the most capable model on the market. It was the one responders could run on their own infrastructure, keeping the work inside their environment. AI Systems Work https://the404.us/field-notes/when-work-stops-looking-like-a-conversation/ https://the404.us/field-notes/when-work-stops-looking-like-a-conversation/ Tue, 21 Jul 2026 00:00:00 GMT Agentic work is beginning to have rhythms instead of turns. Draft marks now need to travel upstream from the finished artifact into the run that produced it. AI Work Interfaces https://the404.us/field-notes/the-web-has-two-front-doors/ https://the404.us/field-notes/the-web-has-two-front-doors/ Mon, 20 Jul 2026 00:00:00 GMT As sites begin exposing structured tools for browser agents, the old human-facing page is getting a second entrance. The design question is what the new door is allowed to do. AI Web Interfaces https://the404.us/field-notes/useful-agent-has-blinders/ https://the404.us/field-notes/useful-agent-has-blinders/ Sun, 19 Jul 2026 00:00:00 GMT As workplace agents gain more access, the better design question is not how much context they can swallow. It is what they are mercifully kept from seeing. AI Work Interfaces https://the404.us/field-notes/when-the-repository-starts-keeping-score/ https://the404.us/field-notes/when-the-repository-starts-keeping-score/ Sat, 18 Jul 2026 00:00:00 GMT Repository-level AI metrics can make agent work easier to see. They can also teach organizations to mistake visible motion for software health. AI Work Quality https://the404.us/field-notes/the-cheapest-model-is-a-management-decision/ https://the404.us/field-notes/the-cheapest-model-is-a-management-decision/ Wed, 15 Jul 2026 00:00:00 GMT As AI products begin routing work across models, providers, budgets, and fallback paths, the humane question is whether people can see what kind of intelligence their work was assigned. AI Work Design https://the404.us/field-notes/personal-ai-needs-an-edit-button/ https://the404.us/field-notes/personal-ai-needs-an-edit-button/ Tue, 14 Jul 2026 00:00:00 GMT Memory-enabled AI is starting to build working theories of its users. The humane design question is whether people can inspect, revise, and expire those theories without becoming administrators of their own personalities. AI Design Life https://the404.us/field-notes/polished-work-needs-draft-marks/ https://the404.us/field-notes/polished-work-needs-draft-marks/ Mon, 13 Jul 2026 00:00:00 GMT As AI agents get better at producing finished-looking decks, documents, spreadsheets, and interfaces, the humane design question is whether review still has somewhere to attach. AI Work Design https://the404.us/field-notes/agent-traffic-needs-house-rules/ https://the404.us/field-notes/agent-traffic-needs-house-rules/ Sun, 12 Jul 2026 00:00:00 GMT As AI agents start visiting the web on our behalf, the humane question is no longer only whether bots may enter. It is what kind of visitor each bot is pretending to be. AI Interfaces Web https://the404.us/field-notes/capability-gates-need-receipts/ https://the404.us/field-notes/capability-gates-need-receipts/ Sat, 11 Jul 2026 00:00:00 GMT As frontier models become more powerful and more restricted, the humane design question is whether access decisions leave enough evidence for people to understand the bargain. AI Trust Work https://the404.us/field-notes/ai-slop-is-scalable-authenticity/ https://the404.us/field-notes/ai-slop-is-scalable-authenticity/ Sat, 11 Jul 2026 00:00:00 GMT AI slop feels uncanny because it scales the polished, repeatable authenticity that platform culture had already taught people to perform. AI Culture Work https://the404.us/field-notes/saved-time-needs-somewhere-to-go/ https://the404.us/field-notes/saved-time-needs-somewhere-to-go/ Fri, 10 Jul 2026 00:00:00 GMT As AI tools begin returning hours to workers, the humane workplace question is what organizations do with that time before it quietly becomes more work. AI Work Design https://the404.us/field-notes/ai-rollouts-need-rehearsal-rooms/ https://the404.us/field-notes/ai-rollouts-need-rehearsal-rooms/ Fri, 10 Jul 2026 00:00:00 GMT As agentic tools move into everyday work, the humane adoption problem is whether people get enough visible practice to learn what good judgment looks like together. AI Work Learning https://the404.us/field-notes/customer-service-ai-needs-an-appeal-path/ https://the404.us/field-notes/customer-service-ai-needs-an-appeal-path/ Tue, 07 Jul 2026 00:00:00 GMT As customer-service agents start resolving, routing, and acting across support systems, the humane interface question is whether people can contest the answer without disappearing into the workflow. AI Work Design https://the404.us/field-notes/ai-capacity-needs-a-civic-interface/ https://the404.us/field-notes/ai-capacity-needs-a-civic-interface/ Mon, 06 Jul 2026 00:00:00 GMT As AI companies turn intelligence into chips, leases, power contracts, and neighborhood infrastructure, the humane design question is whether people can see the capacity they are being asked to live with. AI Design Infrastructure https://the404.us/field-notes/agent-controls-need-weight/ https://the404.us/field-notes/agent-controls-need-weight/ Fri, 03 Jul 2026 00:00:00 GMT As AI work moves from chat into background agents, shortcuts, and physical controls, the humane interface question is whether control feels like responsibility or just another way to go faster. AI Design Work https://the404.us/field-notes/science-agents-need-strange-questions/ https://the404.us/field-notes/science-agents-need-strange-questions/ Thu, 02 Jul 2026 00:00:00 GMT As AI research workbenches start running analyses, querying databases, and producing reproducible artifacts, the humane scientific question is whether they preserve exploration or quietly narrow it. AI Science Work https://the404.us/field-notes/agentic-development-needs-load-bearing-pauses/ https://the404.us/field-notes/agentic-development-needs-load-bearing-pauses/ Wed, 01 Jul 2026 00:00:00 GMT As coding agents keep working after people stop, the surprise infrastructure bill is not only a cost problem. It is a sign that our software delivery systems were designed around human pauses. AI Work Infrastructure https://the404.us/field-notes/institutional-ai-needs-a-training-record/ https://the404.us/field-notes/institutional-ai-needs-a-training-record/ Tue, 30 Jun 2026 00:00:00 GMT As companies turn their own work history into AI capability, the humane design question is whether people can see what the system learned from them. AI Work Design https://the404.us/field-notes/ai-meetings-need-a-guest-list/ https://the404.us/field-notes/ai-meetings-need-a-guest-list/ Tue, 30 Jun 2026 00:00:00 GMT As AI notetakers move from video calls into everyday rooms, the humane design question is whether people know who is listening, what is being kept, and when the room can go back to being a room. AI Work Design https://the404.us/field-notes/ai-work-needs-a-shared-floor/ https://the404.us/field-notes/ai-work-needs-a-shared-floor/ Sat, 27 Jun 2026 00:00:00 GMT As agentic tools become ordinary in some corners of work and exotic in others, the humane design question is how teams keep a shared sense of the work itself. AI Work Design https://the404.us/field-notes/ai-patches-need-maintainers/ https://the404.us/field-notes/ai-patches-need-maintainers/ Fri, 26 Jun 2026 00:00:00 GMT As AI security tools learn to find and draft fixes faster, the humane infrastructure question is whether maintainers gain capacity or inherit a louder queue. AI Security Work https://the404.us/field-notes/agent-work-needs-role-lines/ https://the404.us/field-notes/agent-work-needs-role-lines/ Thu, 25 Jun 2026 00:00:00 GMT As coding agents make technical execution easier to delegate across an organization, the humane design problem is deciding where capability ends and responsibility begins. AI Work Design https://the404.us/field-notes/ai-speed-needs-a-second-clock/ https://the404.us/field-notes/ai-speed-needs-a-second-clock/ Wed, 24 Jun 2026 00:00:00 GMT As AI tools make work feel faster at the keyboard, the humane measurement problem is whether the whole system is actually getting calmer, clearer, and better. AI Work Quality https://the404.us/field-notes/creative-agents-need-rough-edges/ https://the404.us/field-notes/creative-agents-need-rough-edges/ Tue, 23 Jun 2026 00:00:00 GMT As AI moves deeper into design and editing tools, the humane question is whether the work still leaves enough texture for taste to attach. AI Design Work https://the404.us/field-notes/health-ai-needs-a-handoff/ https://the404.us/field-notes/health-ai-needs-a-handoff/ Mon, 22 Jun 2026 00:00:00 GMT As medical models become more capable, the humane design problem is not only whether they can find the right answer. It is how uncertainty, context, and responsibility move back to human care. AI Health Design https://the404.us/field-notes/ai-classrooms-need-a-syllabus/ https://the404.us/field-notes/ai-classrooms-need-a-syllabus/ Sun, 21 Jun 2026 00:00:00 GMT As study bots and teacher copilots move into schoolwork, the humane question is not only whether students used AI. It is whether anyone knows what kind of learning the tool was supposed to leave behind. AI Education Design https://the404.us/field-notes/workspaces-need-housekeeping/ https://the404.us/field-notes/workspaces-need-housekeeping/ Sat, 20 Jun 2026 00:00:00 GMT As agents move into real files, tickets, calendars, and repos, the hidden work is not only giving them better tools. It is making the work itself legible enough to share. AI Work Design https://the404.us/field-notes/local-ai-needs-a-front-door/ https://the404.us/field-notes/local-ai-needs-a-front-door/ Fri, 19 Jun 2026 00:00:00 GMT As AI moves into browsers, operating systems, and everyday apps, running locally is useful. It is not the same as knowing who is inside the room. AI Interfaces Privacy https://the404.us/field-notes/scheduled-ai-needs-a-shelf-life/ https://the404.us/field-notes/scheduled-ai-needs-a-shelf-life/ Thu, 18 Jun 2026 00:00:00 GMT As assistants learn to watch, remind, and run in the background, the humane interface question is not how often AI can return. It is when delegated attention should expire. AI Attention Design https://the404.us/field-notes/agent-actions-need-repair-paths/ https://the404.us/field-notes/agent-actions-need-repair-paths/ Wed, 17 Jun 2026 00:00:00 GMT As AI agents move from suggestions into work systems, the humane interface problem is no longer only approval. It is how people recover when a plausible action has already changed the world. AI Design Work https://the404.us/field-notes/junior-work-needs-a-floor/ https://the404.us/field-notes/junior-work-needs-a-floor/ Tue, 16 Jun 2026 00:00:00 GMT AI may remove some of the routine work that used to define early-career jobs. The humane question is what replaces it as a place to practice judgment before judgment is demanded. AI Work Learning https://the404.us/field-notes/agent-work-needs-stop-conditions/ https://the404.us/field-notes/agent-work-needs-stop-conditions/ Tue, 16 Jun 2026 00:00:00 GMT Coding agents do not only answer prompts. They inherit a workspace, a tool belt, a loop, and sometimes a very loose idea of when the work is over. AI Engineering Work https://the404.us/field-notes/ai-needs-a-degraded-mode/ https://the404.us/field-notes/ai-needs-a-degraded-mode/ Mon, 15 Jun 2026 00:00:00 GMT When a model is rerouted, restricted, or suddenly unavailable, the humane fallback is not to keep the interface looking normal. It is to help people understand what changed and keep the work intact. AI Systems Work https://the404.us/field-notes/reproduction-needs-skepticism/ https://the404.us/field-notes/reproduction-needs-skepticism/ Sun, 14 Jun 2026 00:00:00 GMT AI agents are getting better at rerunning published research. The harder question is whether they can help institutions challenge a result instead of merely producing it again. AI Science Quality https://the404.us/field-notes/public-input-needs-somewhere-to-go/ https://the404.us/field-notes/public-input-needs-somewhere-to-go/ Sat, 13 Jun 2026 00:00:00 GMT AI companies are getting better at asking people what they fear and hope for. Listening only becomes accountability when the public can see what changed and who still has the power to say no. AI Society Trust https://the404.us/field-notes/ai-workflows-have-scar-tissue/ https://the404.us/field-notes/ai-workflows-have-scar-tissue/ Fri, 12 Jun 2026 00:00:00 GMT A prompt stack is often a record of old failures, anxious fixes, and institutional memory. Smarter models do not erase that history; they make us decide what still deserves to survive. AI Work Systems https://the404.us/field-notes/ai-needs-a-nutrition-label/ https://the404.us/field-notes/ai-needs-a-nutrition-label/ Fri, 12 Jun 2026 00:00:00 GMT As AI-generated work moves from answers into actions, people need a small, visible way to inspect what the work depends on before they trust it. AI Interfaces Trust https://the404.us/field-notes/shopping-agents-need-a-receipt/ https://the404.us/field-notes/shopping-agents-need-a-receipt/ Thu, 11 Jun 2026 00:00:00 GMT As AI agents move from finding things to buying them, checkout has to preserve consent, regret, and responsibility after the button disappears. AI Interfaces Work https://the404.us/field-notes/job-search-needs-a-human-shape/ https://the404.us/field-notes/job-search-needs-a-human-shape/ Wed, 10 Jun 2026 00:00:00 GMT As AI starts helping people find jobs, rewrite resumes, and reorganize work itself, the humane question is not whether the search gets faster. It is whether anyone can still be recognized inside it. AI Work Interfaces https://the404.us/field-notes/ai-needs-a-good-logout/ https://the404.us/field-notes/ai-needs-a-good-logout/ Tue, 09 Jun 2026 00:00:00 GMT As assistants become persistent agents across phones, browsers, workspaces, and local machines, the humane design problem is not only what they can do. It is how cleanly they can stop. AI Interfaces Work https://the404.us/field-notes/second-opinions-need-standards/ https://the404.us/field-notes/second-opinions-need-standards/ Mon, 08 Jun 2026 00:00:00 GMT As AI tools start asking other AI tools to critique their plans, tests, and output, the humane question is whether the second opinion is defending a real standard or just making uncertainty look managed. AI Quality Work https://the404.us/field-notes/agent-work-needs-handles/ https://the404.us/field-notes/agent-work-needs-handles/ Mon, 08 Jun 2026 00:00:00 GMT As AI work moves out of chat and into canvases, sites, agents windows, and annotated artifacts, the humane design question is whether people can still grab the work where judgment actually happens. AI Interfaces Work https://the404.us/field-notes/agent-plans-need-witnesses/ https://the404.us/field-notes/agent-plans-need-witnesses/ Fri, 05 Jun 2026 00:00:00 GMT As AI agents learn to plan before they act, the humane design problem is not making the plan look authoritative. It is making sure somebody can still notice what the plan leaves out. AI Work Design https://the404.us/field-notes/advice-bots-need-doubt/ https://the404.us/field-notes/advice-bots-need-doubt/ Thu, 04 Jun 2026 00:00:00 GMT As people ask AI systems what to do with their jobs, bodies, money, children, and relationships, the humane design problem is not better confidence. It is better doubt. AI Health Interfaces https://the404.us/field-notes/the-office-graph-needs-manners/ https://the404.us/field-notes/the-office-graph-needs-manners/ Wed, 03 Jun 2026 00:00:00 GMT As AI tools learn from the relationships, files, meetings, messages, and permissions around work, the humane design question is no longer only what context they can see. It is how politely they use it. AI Work Design https://the404.us/field-notes/disposable-software-needs-aftercare/ https://the404.us/field-notes/disposable-software-needs-aftercare/ Tue, 02 Jun 2026 00:00:00 GMT As coding agents move into knowledge work, more teams will make small tools for small moments. The hard part is deciding what deserves to keep existing. AI Work Software https://the404.us/field-notes/agents-need-badges/ https://the404.us/field-notes/agents-need-badges/ Tue, 02 Jun 2026 00:00:00 GMT As AI agents get identities, permissions, sponsors, and audit trails, the important design question is not only what they can do. It is how they become legible members of the workplace. AI Work Trust https://the404.us/field-notes/people-are-not-exception-queues/ https://the404.us/field-notes/people-are-not-exception-queues/ Mon, 01 Jun 2026 00:00:00 GMT As agents take on more execution, the humane workplace question is not whether humans stay in the loop. It is what kind of loop we are asking them to live inside. AI Work Design https://the404.us/field-notes/documents-became-agentic-workrooms/ https://the404.us/field-notes/documents-became-agentic-workrooms/ Fri, 29 May 2026 00:00:00 GMT As AI moves from answering beside our files to acting inside them, documents, spreadsheets, and decks are becoming rooms where judgment, authority, and evidence have to coexist. AI Work Interfaces https://the404.us/field-notes/ai-memory-needs-edges/ https://the404.us/field-notes/ai-memory-needs-edges/ Fri, 29 May 2026 00:00:00 GMT As assistants remember across chats, projects, repositories, and connected tools, the humane design question is no longer whether memory helps. It is where memory should stop. AI Interfaces Trust https://the404.us/field-notes/search-needs-exits/ https://the404.us/field-notes/search-needs-exits/ Wed, 27 May 2026 00:00:00 GMT As AI search becomes a place where answers, interfaces, planning, and transactions happen, the humane design question is whether people can still leave the answer cleanly. AI Interfaces Knowledge https://the404.us/field-notes/sandboxes-are-the-agent-interface/ https://the404.us/field-notes/sandboxes-are-the-agent-interface/ Tue, 26 May 2026 00:00:00 GMT As agents move from chat into managed work environments, the most important interface may be the room we let them work inside. AI Interfaces Trust https://the404.us/field-notes/provenance-became-part-of-the-interface/ https://the404.us/field-notes/provenance-became-part-of-the-interface/ Tue, 26 May 2026 00:00:00 GMT As AI agents write code, sign commits, and leave traces through our tools, authorship stops being a footer. It becomes part of whether work can be trusted. AI Trust Work https://the404.us/field-notes/instructions-became-infrastructure/ https://the404.us/field-notes/instructions-became-infrastructure/ Tue, 26 May 2026 00:00:00 GMT As agents begin to run from Markdown files, skills, goals, and workflow rules, the quality of ordinary instructions starts to matter like software architecture. AI Work Infrastructure https://the404.us/field-notes/good-tech-ideas-can-come-from-anybody/ https://the404.us/field-notes/good-tech-ideas-can-come-from-anybody/ Tue, 26 May 2026 00:00:00 GMT AI tools may not make everyone a software engineer. Their more interesting promise is that they let more people bring a serious idea close enough to software to be heard. AI Software Creativity https://the404.us/field-notes/ai-code-review-needs-disagreement/ https://the404.us/field-notes/ai-code-review-needs-disagreement/ Mon, 25 May 2026 00:00:00 GMT As AI review moves from comments into suggested fixes and agentic repair, the important question is whether teams still have enough friction to notice what quality actually requires. AI Coding Quality https://the404.us/field-notes/background-agents-need-endings/ https://the404.us/field-notes/background-agents-need-endings/ Fri, 22 May 2026 00:00:00 GMT As AI work moves from active sessions into phones, schedules, locked computers, and remote agent threads, the humane design problem is no longer only how work begins. It is how it stops. AI Work Attention https://the404.us/field-notes/agent-readable-software-needs-caretakers/ https://the404.us/field-notes/agent-readable-software-needs-caretakers/ Tue, 19 May 2026 00:00:00 GMT As software gets redesigned for agents as well as humans, the hidden work is not making everything autonomous. It is keeping systems legible enough to trust. AI Interfaces Maintenance https://the404.us/field-notes/ambient-ai-needs-friction/ https://the404.us/field-notes/ambient-ai-needs-friction/ Mon, 18 May 2026 00:00:00 GMT As AI moves into default models, keyboards, cursors, widgets, and operating systems, the humane design problem is no longer access. It is knowing when help should slow itself down. AI Interfaces Agency https://the404.us/field-notes/ai-is-thinning-the-apprenticeship-layer/ https://the404.us/field-notes/ai-is-thinning-the-apprenticeship-layer/ Mon, 18 May 2026 00:00:00 GMT The first labor problem may not be that AI eliminates entire professions. It may be that it quietly removes the messy early work where people learn how judgment is made. AI Work Learning https://the404.us/field-notes/work-chat-became-an-execution-surface/ https://the404.us/field-notes/work-chat-became-an-execution-surface/ Sun, 17 May 2026 00:00:00 GMT As agents move into Slack channels, schedules, and shared directories, the workplace conversation layer is becoming a place where work gets triggered, reviewed, and quietly delegated. AI Work Interfaces https://the404.us/field-notes/design-code-loop-needs-taste/ https://the404.us/field-notes/design-code-loop-needs-taste/ Fri, 15 May 2026 00:00:00 GMT As AI collapses the distance between canvas and code, the scarce work is no longer translation. It is knowing what should survive the loop. AI Design Quality https://the404.us/field-notes/ai-metrics-are-management-design/ https://the404.us/field-notes/ai-metrics-are-management-design/ Fri, 15 May 2026 00:00:00 GMT As AI work becomes metered by team, model, feature, and cost, the question is no longer whether people use the tools. It is what the organization teaches them to value. AI Work Quality https://the404.us/field-notes/deep-research-changed-the-shape-of-knowledge-work/ https://the404.us/field-notes/deep-research-changed-the-shape-of-knowledge-work/ Thu, 14 May 2026 00:00:00 GMT The important shift is not that AI can summarize faster. It is that research is starting to look like delegated labor instead of assisted search. AI Research Work https://the404.us/field-notes/mcp-made-ai-tools-less-lonely/ https://the404.us/field-notes/mcp-made-ai-tools-less-lonely/ Wed, 13 May 2026 00:00:00 GMT The breakthrough is not one more model. It is a shared way for agents to reach tools, data, and systems without every product pretending to be a closed universe. AI Tools Infrastructure https://the404.us/field-notes/context-engineering-is-replacing-prompt-engineering/ https://the404.us/field-notes/context-engineering-is-replacing-prompt-engineering/ Tue, 12 May 2026 00:00:00 GMT The frontier is moving away from clever wording and toward the design of what the model can see, retrieve, compress, and ignore. AI Design Systems https://the404.us/field-notes/plugins-are-becoming-the-real-ai-product/ https://the404.us/field-notes/plugins-are-becoming-the-real-ai-product/ Mon, 11 May 2026 00:00:00 GMT The model may still be the engine, but the product advantage is moving toward packaged capability: tools, rules, workflows, and embedded know-how. AI Product Workflows https://the404.us/field-notes/the-browser-is-turning-into-an-agent-workbench/ https://the404.us/field-notes/the-browser-is-turning-into-an-agent-workbench/ Sun, 10 May 2026 00:00:00 GMT The browser used to be where we did the work. Increasingly it is becoming the place where we supervise work being done on our behalf. AI Web Attention https://the404.us/field-notes/multimodal-memory-will-change-what-search-feels-like/ https://the404.us/field-notes/multimodal-memory-will-change-what-search-feels-like/ Sat, 09 May 2026 00:00:00 GMT When text, images, audio, and video can live in the same retrieval space, search stops being a library query and starts becoming a memory system. AI Search Media https://the404.us/field-notes/coding-agents-made-management-visible/ https://the404.us/field-notes/coding-agents-made-management-visible/ Fri, 08 May 2026 00:00:00 GMT The real bottleneck in software was never just typing speed. It was the hidden work of scope, delegation, review, context, and trust. AI Coding Management https://the404.us/field-notes/model-choice-became-an-interface-decision/ https://the404.us/field-notes/model-choice-became-an-interface-decision/ Thu, 07 May 2026 00:00:00 GMT When multiple frontier agents live inside the same product, model selection stops feeling like loyalty and starts feeling like workflow design. AI Product Strategy https://the404.us/field-notes/ai-capability-is-outpacing-workplace-theory/ https://the404.us/field-notes/ai-capability-is-outpacing-workplace-theory/ Wed, 06 May 2026 00:00:00 GMT The tools are getting better at acting, researching, coding, and remembering, while many organizations still think work is a matter of attendance, output, and managerial visibility. AI Work Health https://the404.us/field-notes/quality-is-the-next-ai-feature/ https://the404.us/field-notes/quality-is-the-next-ai-feature/ Tue, 05 May 2026 00:00:00 GMT After the fascination with speed, the durable differentiator may be whether a tool helps people produce work that is calmer, cleaner, and more worth trusting. AI Quality Society https://the404.us/field-notes/problem-with-okrs/ https://the404.us/field-notes/problem-with-okrs/ Thu, 09 Apr 2026 00:00:00 GMT OKRs can name a destination, but they often miss the conditions that make health, judgment, and quality possible in the first place. Ecology Work Emergence https://the404.us/field-notes/ai-morning-routine/ https://the404.us/field-notes/ai-morning-routine/ Thu, 19 Mar 2026 00:00:00 GMT A useful routine generated by a model can still fail on contact with energy levels, mood, weather, and the older operating system known as a body. AI Work Reflection https://the404.us/field-notes/weight-loss-systems/ https://the404.us/field-notes/weight-loss-systems/ Thu, 12 Mar 2026 00:00:00 GMT The shift was not toward harsher discipline, but toward better systems, fewer identity dramas, and more attention to what my days were already teaching me. Health Mindfulness Systems https://the404.us/field-notes/delight-systems/ https://the404.us/field-notes/delight-systems/ Thu, 05 Mar 2026 00:00:00 GMT Scarcity, anticipation, pacing, and surprise are not side effects. They are the architecture of how systems become memorable. Games Travel Design