The Broad Way
[ Sharp Mind · Sharp Blade · Sharp Spirit ]
A new release needs to solve a difficulty before taking over as my default. The chart can look excellent while yesterday's configuration remains suitable. I want a work-related reason to switch, espec...
Choosing the lowest model price before defining the deliverable saves money on the wrong variable. I want an inexpensive option that finishes the job with sufficient quality. Paying for the most capab...
A specialized tool that makes its answer harder to check adds work for the person it is supposed to help. Fluency looks good in a demonstration. In use, I want to return to the source and correct a pr...
Coordination rules age, and I do not want a new model trapped inside a workaround for an old problem. Writing instructions takes effort. That does not give them a permanent claim on every assignment t...
A better model needs to improve the work that reaches me. I like seeing capability advance, but I get impatient when the tool leaves that improvement stuck behind a poor integration. The promise of co...
The cost of leaving an API belongs in the decision before its shutdown notice arrives. Useful features on day one do not pay for rebuilding an integration. When the calendar forces a switch, whoever d...
I want to pay for a completed task, and a token discount alone cannot tell me what that costs. A cheaper call can still fund an endless meeting between agents. The calculation needs to reach a deliver...
Claude's watermark does little to reconstruct work shared across several tools. I want to know how a proposal became an accepted change. An estimate of model participation answers a much smaller quest...
A random maintainer should never have to contain an agent that escaped the scope of an evaluation. That is what bothers me about the AISI report: somebody outside the experiment had to recognize and r...
An agent failure report without the test setup is about as useful to me as a benchmark that never names the machine. You can argue almost anything from it. I want to know what the agent attempted and ...
A stuck task deserves another approach before it gets another day of my patience. On July 13, 2026, I reported a case where an agent resolved in minutes something that had been taking all day. It was ...
An agent busy for hours interests me when I return and find that work has advanced. On July 9, 2026, OpenAI introduced ChatGPT Work for long tasks across applications and files. That promise of contin...
A coordinator that repeats what each agent said leaves me all the work of joining the deliveries. I can collect messages myself, with coffee and an alarming number of tabs. When I ask for coordination...
Dumping every workstream's context into each agent is a poor way to delegate. On June 17, 2026, I objected to exactly that idea. I want agents collaborating, and I still want to know which job each on...
A workflow that loses task state when it loses model access is poorly prepared to continue working. On June 12, 2026, Anthropic announced a global suspension of Fable 5 and Mythos 5 to comply with an ...
A rule that adds ceremony needs to get out of an agent's way, including a rule I wrote myself. It is easy to complain about a slow model while treating my configuration as sacred. If I asked it to req...
A proactive agent needs to know when the work is finished, or I get a subscription to suggestions I never requested. The Gemini Spark announcement in May 2026 presented continuous tasks and proactive ...
A skill pointing to a dead API needs maintenance, however polished its description looks. OpenAI recorded the removal of DALL·E 2 and DALL·E 3 from its API in May 2026. The endpoint leaves the catalog...
I want parallel work that finishes tasks; spare agents give me more things to follow. In June 2026, I asked for persistent agents to gain speed. The next day, I suspected there were too many and sugge...
I prefer an AI that finds the right context to one that drags my entire history into every answer. More memory also means more old decisions available to appear in the wrong place. I want to know why ...
Another desktop assistant needs to reduce the effort of resuming a task. If it merely adds somewhere else to search for the right conversation, I have acquired an organization problem with a nice icon...
I want a review that explains a concrete problem, because a vague second opinion only adds to my reading queue. In September 2026, I described my Claude arrangement: Sonnet implemented, Opus reviewed....
I want the best model I can use, and I am not leaving work waiting for an inaccessible name. Mythos Preview makes that distinction clear. Anthropic disclosed the model in April 2026 with cybersecurity...
Sending a task from my phone is worthwhile only if the agent does its share of finding context. If I have to prepare a dossier on a tiny screen, the convenience is gone before work begins. I want to s...
OpenAI just acquired Astral. If that name doesn't ring a bell, the tools will: ruff (the Python linter that made flake8 look like it was running on a TI-84) and uv (the package manager that finally ma...
Cursor just announced 1 million daily active users and fifty thousand businesses, including Stripe, Figma, and Shopify. It has a fresh 900M dollar valuation and is building its own model. Meanwhile, I...
I'm sitting in Vitoria, Brazil, 5,000 miles from Washington D.C., reading about the White House preparing to send Congress a national AI regulatory framework. And I can already feel it in my codebase....
Every vendor on Earth just discovered AI agents. Huawei launched AgentArts. Google Cloud published "AI Agent Trends 2026." Salesforce has "Agentforce." ServiceNow has "AI Agents." Microsoft has "Copil...
Got tagged in an issue thread yesterday where someone was demanding a fix for a bug in a library that is free, open source, maintained by one person in their spare time, and used by this complainer in...
Had a call with a startup last week. "Our search is really slow, we are thinking about implementing AI-powered semantic search with vector embeddings." How many records? "About fifty thousand." They...
Everything has a chat interface now. Customer support? Chat. Banking? Chat. Ordering food? Chat. Filing taxes? Believe it or not, also chat. I tried to change my flight the other day. Instead of a bu...
"I don't like this variable name." A personality disorder wearing a GitHub avatar has entered the code review. Code review should check architecture, logic, and security. Architecture: does this chan...
They show up in every architecture thread on the internet. Someone posts a nuanced question about database design, event sourcing, or multi-tenant data isolation, and within three minutes comes the co...
If all your services deploy together, share a database, and cannot function independently, congratulations. You built a monolith with network latency. You made everything worse and called it architect...
Every six months, someone on Twitter reignites the monorepo versus polyrepo war and I lose another year of my life watching people argue about something that has no universal answer. "Monorepos are t...
Cheaper tokens excite me less when the work stays stuck. NVIDIA announced Rubin in January 2026, promising substantially lower inference costs than Blackwell, with partner products expected in the sec...
You want someone who can architect distributed systems, design a CI/CD pipeline from scratch, mentor three juniors, and own the entire payments infrastructure. Cool. The budget says "junior with React...
What went well AI coding tools went from "cute toy" to "I genuinely cannot imagine working without this." Claude Code became my pair programmer. I can describe a complex migration across 42 NestJS mo...
Every six months someone publishes a blog post titled "JavaScript Fatigue Is Killing Me" and it goes viral because developers love agreeing that things are hard. I think "JavaScript fatigue" is a pro...
Your CRUD API does not need zero-cost abstractions, and your Next.js landing page does not need memory safety guarantees. The todo app that talks to Supabase and renders some cards will be fine withou...
The moment a senior developer hits 5 years of experience, they build a blog. Shipping a product, contributing to open source, or mentoring a junior can apparently wait. They build it from scratch wit...
It's 2025. We were promised flying cars, or at least a cure for the common cold. Instead, we got LLMs that can write Shakespearean sonnets about toaster ovens but can't center a div without hallucinat...