标签：#AR

There are already many posts about what agent skills are and how to create your own, so in this post I want to dive into the various skills I use to assist in development. The Skills Grill Me Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me". I start every larger task with this excellent skill created by Matt Pocock. I either start this with an already prepared PRD / detailed task description or use it for discovery purposes. The agent will then ask many questions to align language and functional requirements, so fewer hallucinations happen in follow up requests. You should be well equipped to answer the agent's question or the grill me session can go on for a long time. I had it ask me way over 50 questions when not answering detailed enough. As a little extra I added an extra request to the skill to prompt me if I want to create the PRD when the alignment phase is over, this leads us to the next skill. To PRD Turn the current conversation context into a PRD. Use when user wants to create a PRD from the current context. This will simply take the current conversation and creates a PRD out of it, we do this to summarize the conversation so we can easily start a new context window with all information present To Issue Break a plan, spec, or PRD into independently-grabbable GitHub issues using tracer-bullet vertical slices. Use when user wants to convert a plan into issues, create implementation tickets, or break down work into issues. Another excellent skill by Matt Pocock. I modified the skill slightly to use the GitHub MCP to create issues based on a PRD or planning session. But I often found that letting an agent implement those tasks it resulted in a large amount of code and that is why I added the to tasks skill To Task Break down a single GitHub issue into a sequential list of small i

In a bid to dismiss a lawsuit over xAI’s polluting gas turbines, the Justice Department claimed the company is integral to military operations—including the Iran War.

Anthropic leaders flew to Washington, DC, to meet with White House officials on Monday. After high-level talks, they're still split on the risk Claude Fable 5 presents.

Class, Record, and Struct serve as blueprints for creating new objects. Classes Classes are the foundational building blocks of Object-Oriented Programming (OOP). Blueprint and instance // blueprint public class Person { public int ID { get ; set ; } public string Name { get ; set ; } } // instance var p = new Person { ID = 1 , Name = "Mirza" } Inspection Printing the class instance in the console will just display it's name: var p1 = new Person { ID = 1 , Name = "Mirza" }; Console . WriteLine ( p1 ); // Person To print the actual, we'd need print each property explicitly: Console . WriteLine ( p1 . ID ); // 1 Console . WriteLine ( p1 . Name ); // "Mirza" Mutation C# classes are mutable, meaning the originally set values can be altered: var p1 = new Person { ID = 1 , Name = "Mirza" }; Console . WriteLine ( p1 . Name ); // "Mirza" p1 . Name = "Armin" ; Console . WriteLine ( p1 . Name ); // "Armin" That said, this can be tweaked by changing the accessor the class property. public class Person { public int ID { get ; set ; } public string Name { get ; init ; } // <-- } This time around if we try to change the value of the Name property after initialization, the C# compiler will start to complain. var p1 = new Person { ID = 1 , Name = "Mirza" }; p1 . Name = "Edis" ; // ❌ The init accessor allows you to create properties that can only be assigned a value during object initialization. Memory location Classes are reference types and are allocated on the heap. If we create two distinct class objects and assign one to the other, both objects will point to the exact same reference in memory: var p1 = new Person { ID = 1 , Name = "Mirza" }; var p2 = new Person { ID = 2 , Name = "Mirza" }; p1 = p2 ; p1 . Name = "Sead" ; Console . WriteLine ( p1 . Name ); // Sead Console . WriteLine ( p2 . Name ); // Sead If we change a value in one of the objects, it will be reflected in both. Equality Two class instances (objects) aren't equal even if they share the same values: var p1 = new P

At Inithouse — a studio running parallel product experiments — we built Be Recommended , a tool that checks how visible your brand is across ChatGPT, Perplexity, Claude, and Gemini. The idea sounded simple: query multiple AI models, score the results, show a report. It was not simple. Here are four technical mistakes we made shipping v1 — and the fixes that actually survived production. Mistake 1: Rate Limiting Was an Afterthought We treated rate limits as edge cases. They were not. Every AI provider has different rate-limit headers, different backoff expectations, and different definitions of "too many requests." Our first architecture just retried on 429. That turned a rate limit into a cascade — one provider throttling triggered a retry storm that cascaded to the others. The fix: Per-provider circuit breakers with exponential backoff. Each provider gets its own state machine. When a circuit opens, we serve cached results for that provider and mark the score as "partial" in the UI. Users see real data, not a spinner that never resolves. At Audit Vibe Coding — another tool in our portfolio focused on code quality audits — we observed the same pattern in a different domain: external API dependencies need isolation. The lesson transferred directly. Mistake 2: The Caching Strategy Was Too Naive Our first cache key was query + model . That breaks immediately — AI model responses drift over time, and a cached result from two weeks ago is misleading. We also had no invalidation strategy beyond TTL. The fix: Cache by query + model + week_number . Weekly invalidation with stale-while-revalidate: serve the cached score instantly, trigger a background refresh, update the display when new data arrives. Users get instant feedback and fresh data within the same session. We measured the impact across our portfolio: stale-while-revalidate cut perceived load time from 8+ seconds to under 1 second for returning visitors. The background refresh means scores stay current without the

OpenAI introduces Deployment Simulation, a method to predict AI model behavior before deployment using real conversation data to improve safety and evaluation accuracy.

AI is once again at the heart of a college graduation protest — this time for the technology's use in Google's defense contracts.

Um relato honesto de alguém que trabalha com design, vive com TDAH e está cansada de dicas genéricas Tem um tipo de artigo sobre organização que eu já sei de cor. É sempre alguma variação de: “faça uma lista, use Pomodoro, durma 8 horas e beba água”. Só que tem um cenário que quase nunca aparece nessas listas: O momento em que você não é neurotípica, está em burnout, tem duas tarefas importantes com o mesmo prazo e nenhuma técnica milagrosa resolve. É sobre isso que eu quero falar aqui. Sumário: O cenário caótico (e bem real) Por que o Pomodoro não funciona pra todo mundo Burnout em quem tem TDAH O dia em que duas tarefas importantes têm o mesmo prazo Estratégia 1: uma prioridade verdadeira por dia Estratégia 2: subtarefas em vez de cronômetro Estratégia 3: time blocking gentil (agenda que não te esmaga) Estratégia 4: reduzir fricção em vez de exigir mais disciplina Estratégia 5: contratos curtos consigo mesma E quando nada disso parece suficiente? Referências O cenário caótico (e bem real) Imagina o seguinte: Projeto A : entrega do pitch da pós, com prazo na sexta. Projeto B: preparar apresentação do roadmap, também para sexta. Você já está cansada, a cabeça rodando, o corpo em modo economia de energia. Aí você joga no Google “como se organizar” e recebe de volta: “Use a técnica Pomodoro, 25 minutos de foco, 5 de pausa.” E você pensa: “Amiga, eu mal estou levantando da cama. Você quer que eu vire um cronômetro humano?” A real é que muita técnica de produtividade tradicional foi pensada para cérebros neurotípicos. Quando a gente vive com TDAH, burnout ou os dois juntos, essa lógica simplesmente não encaixa tão bem. Por que o Pomodoro não funciona pra todo mundo Pomodoro é ótimo… para algumas pessoas. Mas tem motivos bem específicos para ser um caos para muitos de nós. Por exemplo: A pausa obrigatória, interrompe justo quando o foco finalmente chegou. A sensação do timer contando, aumenta a ansiedade em vez de ajudar. Cada “reinício de ciclo” vira mais uma micro deci

You scan one to pay at a sari-sari store, pull up a restaurant menu, or board a flight. The QR code has quietly become one of the most universal pieces of interface design on the planet. But it was never meant for any of that. The QR code was invented in 1994 to solve a very specific problem on a Japanese car factory floor, and the engineering decisions made under that constraint are exactly why it later conquered the world. A barcode problem on the assembly line In the early 1990s, Toyota's manufacturing arm had a data problem. Tracking thousands of distinct components through production meant scanning barcodes, and barcodes are stingy: a standard one-dimensional barcode holds roughly 20 characters. Workers were ending up with parts plastered in ten or more barcodes just to encode enough information, and each one had to be scanned separately. It was slow, and on an assembly line, slow is expensive. Masahiro Hara, an engineer at Denso Wave, a Toyota subsidiary, took on the challenge of designing something better. He wanted a code that could hold far more data, be read much faster, and tolerate the dirt, smudges, and odd angles of a real factory rather than a clean lab. Designing for speed and any angle The breakthrough was going two-dimensional. By encoding data in a grid of black and white squares rather than a single row of lines, Hara's team could pack in thousands of characters instead of a few dozen. The name they chose, QR for "Quick Response," was a direct promise about scanning speed. The most recognizable feature of a QR code, the three large squares in its corners, solves the hardest part of the problem: letting a scanner instantly find the code and work out its orientation no matter how the part is turned. Hara's team analyzed printed material to find a black-and-white sequence that almost never occurs naturally in text and images, and settled on a ratio of 1:1:3:1:1 for those corner markers. Because that pattern is so rare in everyday print, a scanner ca

Despite continued benefits, anti-vaccine rhetoric has driven down vaccination.

Swap out your creaky old box fan for a new model that lights up, mists, or even follows you around the room.

Don’t be fooled by the compact size of this soundbar. It’s a solid option for smaller TVs or spaces without having to sacrifice sound quality.

This is an adapted excerpt from a guide in my AI Knowledge Hub. The full interactive version is linked at the end. Agentic design patterns are named control structures for arranging LLM calls and tools. This post gives you the decision rule for picking one, the exact shape of each pattern, and the cost each adds — so you can match a task to the minimum structure that solves it. Everything here is model-agnostic and grounded in Anthropic's Building Effective Agents and the Claude Agent SDK. Workflow vs. agent: the split that decides everything Anthropic divides all agentic systems into two categories, and the split decides every downstream tradeoff: Category Definition Control lives in Use when Workflow LLMs and tools orchestrated through predefined code paths Your code You can pre-map the decision tree; want accuracy, control, lower cost Agent LLMs dynamically direct their own processes and tool usage , maintaining control over how they accomplish tasks The model Open-ended task where you can't predict the number of steps Every pattern composes one unit: the augmented LLM — an LLM enhanced with retrieval, tools, and memory. It generates its own search queries, selects tools, and decides what to retain. If a single augmented LLM call solves the task, stop — no pattern required. The escalation rule is the whole game: find "the simplest solution possible, and only increasing complexity when needed" — which "might mean not building agentic systems at all." Agentic systems trade latency and cost for better task performance, so only escalate when a specific failure mode forces it. The agent loop: gather → act → verify → repeat For open-ended tasks, every agent runs the same four-beat loop: Gather context — read files, run agentic search ( grep / find / tail to pull relevant slices instead of whole files), or delegate to subagents with isolated context windows. Take action — execute via tools: bash, code generation, file edits, MCP servers. Verify work — check the result b

Anyone who's wired an autonomous agent into email and calendar the traditional way knows the identifier sprawl: an OAuth client ID, a refresh token per user, a Gmail-specific message ID format, a Microsoft Graph calendar ID, and a webhook subscription ID for each — all with different lifetimes, all able to break independently. Half your "integration" code is really identifier bookkeeping. Nylas Agent Accounts collapse all of that into one value. When you create an account (the feature's in beta), the response hands you a grant_id , and that single string is the handle for everything the agent does — mail, calendar, contacts, attachments, and the webhooks reporting on all of them. One ID, the whole surface Every operation addresses the same path family, /v3/grants/{grant_id}/* : POST /v3/grants/{grant_id}/messages/send # send mail GET /v3/grants/{grant_id}/messages # read the inbox GET /v3/grants/{grant_id}/threads/{thread_id} # full conversation GET /v3/grants/{grant_id}/attachments/{id}/download POST /v3/grants/{grant_id}/events # host a meeting POST /v3/grants/{grant_id}/events/{id}/send-rsvp # respond to one GET /v3/grants/{grant_id}/contacts GET /v3/grants/{grant_id}/rule-evaluations # audit trail This isn't an abstraction invented for agents — an Agent Account is literally just another grant, the same primitive used for connected Gmail and Outlook accounts. The supported endpoints reference puts it as "same endpoints, same auth, same payloads." Anything you built for connected accounts works against an agent's grant unchanged. And the resources behind the ID are real: six system folders provisioned automatically ( inbox , sent , drafts , trash , junk , archive ), a primary calendar that speaks standard iCalendar, and outbound messages capped at 40 MB total. Webhooks route by it too The inbound side completes the picture. You subscribe once at the application level, and every notification — message.created , event.updated , message.bounce_detected , and the rest

Week one: a single test mailbox on a trial domain, provisioned by hand from the dashboard. Week twelve: a fleet of agent mailboxes spread across customer domains, each sending real mail with its own quota and reputation. The API calls are the same at both scales — what changes is everything around them: how you provision, how you share configuration, and how you keep one bad sender from pausing the fleet. Here's what the path from one to thousands looks like with Nylas Agent Accounts , which are currently in beta. Provisioning is a loop, not a ceremony There's no OAuth dance to scale around. Creating a mailbox is one POST with "provider": "nylas" — no refresh token, no consent screen — so a fleet provisioner is just iteration: curl --request POST \ --url "https://api.us.nylas.com/v3/connect/custom" \ --header "Authorization: Bearer <NYLAS_API_KEY>" \ --header "Content-Type: application/json" \ --data '{ "provider": "nylas", "workspace_id": "<WORKSPACE_ID>", "settings": { "email": "agent-0042@agents.yourcompany.com" } }' Two scaling-relevant details in that request. First, the domain: one application can manage accounts across any number of registered domains, and the docs explicitly recommend splitting high-volume outbound across multiple domains ( sales-a.yourcompany.com , sales-b.yourcompany.com ) so reputation damage on one doesn't contaminate the rest. Second, the workspace_id : passing it at creation is how each account picks up its configuration, which brings us to the part that makes fleets manageable. Configure workspaces, not grants At fleet scale, per-grant configuration is a non-starter. The model here is indirection: policies and rules attach to workspaces , and every account in a workspace inherits them. One policy object — daily send quota, storage cap, retention windows, spam sensitivity — governs a thousand mailboxes, and updating it updates all of them at once. The recommended carve-up is one workspace per agent archetype: outreach agents get a work

Every Agent Account you create lands in a workspace — the only question is whether you picked it or the platform did: curl --request POST \ --url "https://api.us.nylas.com/v3/connect/custom" \ --header "Authorization: Bearer <NYLAS_API_KEY>" \ --header "Content-Type: application/json" \ --data '{ "provider": "nylas", "workspace_id": "<WORKSPACE_ID>", "settings": { "email": "test@your-application.nylas.email" } }' That top-level workspace_id is optional, and what happens when you omit it is one of the more under-read parts of the Nylas Agent Accounts docs (the feature's in beta, so read with that in mind). Workspaces aren't cosmetic grouping — they're the carrier for every policy limit and mail rule that governs your agents. Getting placement wrong means an agent running with the wrong send quota, the wrong spam settings, or no inbound filtering at all. The placement algorithm When a new account is created, placement resolves in this order: Explicit workspace_id on the request — the account goes exactly where you said. The target workspace must belong to the same application; ownership is validated and mismatches are rejected. Auto-grouping by domain — if a workspace has auto_group enabled and its domain matches the new account's email domain, the account joins it automatically. The default workspace — everything else lands here. That third bucket is the interesting one. What the default workspace actually is Every application gets exactly one default workspace, created and managed by the platform. You can't delete it, and you can only update two of its fields: policy_id and rule_ids . Everything else is managed for you. It's easy to read "default" as "throwaway," but it's better understood as your safety net. Any account you forget to place explicitly — a quick test from the CLI, a provisioning script missing the workspace parameter — inherits whatever the default workspace carries. Which leads to a practical recommendation: attach a conservative baseline policy and

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. When I landed in Seoul after a grueling 12-hour flight from San Francisco, I walked through an unmanned immigration checkpoint, where a machine scanned my face and passport. On the subway home,…

Series Final. The four mechanisms covered across this series — knowledge graph, Auto Review, Self-Healing, Recurrence Prevention — plus the non-engineer-PR application that sits on top of them, all hang off a single conviction: AI isn't something to trust; it's something to design. The 'I don't trust AI to fill in the blanks for me' framing this lives inside isn't doubt about generation quality, but the clear-eyed acceptance that AI has no idea what context wasn't handed to it, and that 'ideal behavior with no spec given' is a fantasy. The starting point goes back to 2025, when I was trying to figure out how to make AI actually understand a large codebase — and ran into walls on both context window scaling (lost in the middle, attention dilution) and learning-based approaches (machine unlearning, destructive interference). GraphRAG + MCP became the way out: hand AI only the facts it needs, when it needs them, so it doesn't have to infer. From code-graph (which I burned two months on and threw away) to the current product-graph (cpg). This piece is the philosophy and the trial-and-error behind the whole series: harnesses confine where hallucinations are allowed to happen, design is translating principles into your own use cases, and Coverage 90% as a solo target breaks the implementation.

A blunt playbook for devs who don’t want to turn into autocomplete zombies. The first time an AI wrote code for me, I felt like I had unlocked cheat codes for real life. I typed a half-baked function name, hit enter, and suddenly I had a block of code that looked legit. It was magical. The second time, though? It suggested something so catastrophic basically the programming equivalent of pulling the fire alarm that I realized: this thing is less “mentor” and more “overconfident intern who thinks they know pointers but actually just broke prod.” That’s where most of us are right now. AI is everywhere: in our IDEs, our docs, even sneaking into PR reviews. Some days it feels like rocket fuel; other days it feels like an autocomplete with a drinking problem. The tricky part isn’t whether AI is “good” or “bad.” The tricky part is how we, as developers, use it without becoming lazy, dependent, or worse complacent. Because here’s the uncomfortable truth: AI won’t replace you, but bad AI habits absolutely will. TLDR : This article is a survival guide for developers in the AI era. We’ll break down why AI feels both magical and mid, the five switches that make AI actually useful, when to trust and when to verify, how to use AI as a research assistant (not a code monkey), the dangers of autocomplete brain, and a playbook for building a healthy workflow. Why AI feels both magical and mid Every dev I know has had that moment with AI. The first time it autocompleted a function and nailed it, you probably thought: “Wow… this thing just saved me half an hour.” It’s the same dopamine hit as discovering ctrl+r in bash or realizing you can pipe grep into less . Pure wizardry. But the honeymoon ends quickly. The same tool that wrote a clean utility function also happily hallucinates imports that don’t exist, invents APIs, and will confidently explain things that are flat-out wrong. It’s like pair programming with someone who sounds senior but has never actually shipped code. The magic-

I had a write that spanned two physically separate databases. A rename that had to propagate across several tables in one database and a couple of tables in another, and the two had to stay consistent. No distributed transaction coordinator was available to me. So I did the obvious thing: opened a transaction on each, did the work, and committed them one after the other inside a try/catch with rollbacks on both sides. It felt safe. It compiled. It passed tests. Then I drew the failure on a whiteboard, and the safety evaporated. The window that ruins everything Here's the structure, simplified: await using var txA = await dbA . Database . BeginTransactionAsync (); await using var txB = await dbB . Database . BeginTransactionAsync (); await DoWorkOnA ( dbA ); await DoWorkOnB ( dbB ); await txA . CommitAsync (); // <-- succeeds await txB . CommitAsync (); // <-- what if this throws? Two transactions do not make one atomic operation. CommitAsync is a point of no return, and there are two of them. Between the first commit returning and the second one starting, there is a window. If txB fails in that window — the connection drops, the process is killed, the database hiccups — then A is permanently committed and B never happens. Your rollback in the catch block is useless: you can't roll back txA , it's already durable. The two databases now disagree, and nothing in your code will heal that on its own. This is the dual-write problem , and it's not a bug you can fix by being more careful with try/catch. The atomicity you want simply isn't available from two independent commits. Ordering them, nesting them, wrapping them — none of it closes the window, because the window is inherent to having two commit points. Why "it's never failed" isn't reassurance The seductive thing about this pattern is that the window is small, so in practice it almost never triggers. You can run it for a year and never see an inconsistency. That's exactly what makes it dangerous: it trains you to tr