I got tired of looking up the syntax for checking what processes are running on the 3000 port, so I created an alias for it:

This was definitely not on my bingo card this week... anthropic.com/news/fable-myt…

I got to see my friend @Snugug at IO (we get to work together now!) and he asked me some questions about WebMCP. I would say I'm sorry about the pun at the end but of course I'm not.

After taking a step back from speaking, I decided to end my sabbatical to talk at RefreshDC about the methodology nobody wants to discuss in the AI era: Documentation-Driven Development. For a recap of the event along with key takeaways, check out: bencodezen.io/blog/refreshdc…

CEOs are quietly realizing the AI replacement plan has a problem. Two problems, actually. One: the token costs for running AI agents are now exceeding what they were paying the employees they fired. Two: when the tokens run out, the AI stops. Just stops. No continuity. No workaround. Just a spinning wheel where your workforce used to be. You fired humans to save money and bought a subscription that bills you into a corner. The employees you let go knew what to do when things broke. The AI just invoices you for the outage. And then there’s the permission problem nobody wants to talk about. To do its job, the AI agent needs access. Full access. Your systems, your patents, your contracts, your future plans. Everything you spent years building, handed over to a process that has no loyalty, no discretion, and no skin in the game. You didn’t hire a replacement. You gave a stranger with no soul the keys to everything you own. Enjoy.

💥 Game changer for Web Development announced at GoogleIO- Modern Web Guidance! It’s expert-vetted skills for web development based on best practices of latest specs and APIs. It ensures your agent/coding harness doesn’t default to older and out of date patterns to build sites.

Workshops at #VueConfUS kick off next Tuesday!🚀 Hands-on learning, real-world problems, & expert guidance for @vuejs devs ready to level up 👀 With David Nahodyl, @danielkelly_io, @bencodezen, @martinrojas & @joemocha Which are you most excited for? 🎟 vueconf.us

Absolutely love this sentiment that my friend @hungsu said recently: "I am generally finding my best results with AI are things where I could have done them myself, eventually."

token merchant suggests using more tokens

📍 I made a new drawing about Context Windows. About “lost in the middle”, how RAG affects it, tokenizers and more. Understanding context windows help you debug and leverage LLMs most effectively. You see why people like Boris from Claude refresh the entire window at times.

Never underestimate how much time and effort you can waste by trying to automate a process you do not understand manually.

This paper about throttling web agents using Reasoning Gates is pretty interesting! For stealth agents that ignore no-crawl directives, they give them multi hop tests of intelligence- It gives the agent very complex logic problems to solve that it would take humans hours of research to do. For an agent, it takes a matter of seconds. The token usage is low for answering once, but too high for multiple attempts. So it introduces computational asymmetry. Getting to an airline site- negligible. But deploying many many agents to perform malicious scraping/DDoS? The server bills alone would bankrupt malicious actors before they got anywhere. arxiv.org/html/2509.01619v1

It's funny to me that the hype train around AI wants the narrative of "you don't need to code anymore" to be a reality. As if anyone who enjoys coding is insane to do so since "AI can do it."

There's definitely something magical about attending these events in-person that doesn't even remotely get captured in virtual events. I have so many good memories of all the past @vueconfus and looking forward to creating many more this year. Hope to see y'all there!

I gave a talk last week and someone asked why we need an agent model for the web- The web needs an established interface layer. WebMCP, though not complete, is part of a strong foundation- it exposes structured endpoints agents can use directly, giving a clearer defined course. I, for one, don’t want agents YOLO’ing around like the wild west. We can build structured well lit paths for better security, rate-limiting, and give observability and control to users.

Always look at incentive structures. On here, this means inflammatory and overly simple statements gain traction- "open source is dead!" or "Software Engineering is dead!" etc A nuanced post exploring tradeoffs and considerations will be drowned out for a hypey quip for debate

Excited to give my first AI talk at @RefreshDC! eventbrite.com/e/tdd-add-ddd…

if your data is stored in a database that a company can freely read and access (i.e. not end-to-end encrypted), the company will eventually update their ToS so they can use your data for AI training — the incentives are too strong to resist

Level up your Vue apps! 🚀 In this hands-on workshop, @bencodezen shows how to build production-grade @Vuejs apps using reusable patterns, automated workflows, and testing best practices for real-world projects. vueconf.us/session?sessionId… 🎟️ Get tickets- vueconf.us

I can't go back to the regular YouTube UI after this 😅 Obsidian Reader now makes the transcript interactive so you can scrub, highlight, auto-scroll. It feels so nice.