Dad, drummer, engineer, co-founder @fullstory.
Atlanta, GA, USA
Joined April 2008
- Tweets436
- Following233
- Followers931
- Likes2.1K
My new favorite QotD:
"You are a software engineer, you literally learned how to make sand think [..,] and now you can't do anything about it?"
something that keeps bothering me is how some people see slop as something that is happening to them, like they're some kind of helpless bystander powerless to do anything about it
excuse me but you are a software engineer, you literally learned how to make sand think and make computers do your bidding?! and now you can't do anything about it? please
the 2 things you should know to cut through all the noise:
1. you need to give your agent a way to verify its own work (i just posted about this so won't rehash)
2. your codebase is memory
the 2nd is not new either, we have always had lint rules, compilers with good diagnostics and strong type systems, LSPs, etc to help us write better code. there is no difference when agents write the code instead, you just need more of them, and stronger ones.
forget rules and skills. you can add those later.
yes, your codebase is a form of memory to agents! agents love extending what already exists. so every nit and every annoyance you've ever seen, every mistake and bad pattern you observe from agents, needs to turn into a constraint, or the agents will just keep repeating those same mistakes like a virus in your code. then you keep observing and repeating this process until your codebase is locked down enough that even agents with little context can one shot a high quality PR.
an even better direction is to refactor/rewrite your code such that agent anti-patterns become categorically impossible. use type systems to your advantage, use libraries/frameworks that reduce the architectural search space, or heck, create your own framework like we did for grok bot and design high quality primitives and well thought out APIs
Also: In the slop-era, writing *less*, effectively, is the real skill. Pascal had some inkling of this in the 18th century:
"Je n’ai fait celle-ci plus longue que parce que je n’ai pas eu le loisir de la faire plus courte."
TFW you let Fable run wild on a research problem, now you have to make sense of a bunch of calculus before you can evaluate whether it even makes sense...
But at least @zeddotdev's delta.dev is making it a hell of a lot easier to manage these research threads!
Every once in a while I feel like younger engineers need a reminder of our illustrious history... of vi/emacs flame wars.
Ed is the standard text editor.
cs.wellesley.edu/~cs249/Reso…
I know I always say this, but we have barely begun to scratch the surface of what's possible. When you find the right capabilities, shape, and skills for agents to GSD, it unlocks so much that was previously a PITA to do by hand.
Headless access to @fullstory has grown 25x from April to September in terms of tool call volume. This is the new way of getting insights from user experience data:
- No dashboard --> Only insights, presented however you want
- No human watching session replay --> Agents watching them for you
- No manual analysis --> Automated loops. Getting stuff done.
Joel Webber retweeted
Once agents fill out all my web forms at least I won't experience bitter ire at custom select's that don't work with my keyboard
Corollary: If you can't make your point briefly and clearly enough that a motivated reader can make sense of it quickly -- you don't understand it well enough to waste their time on it.
LMs are really effective at helping you understand complex outputs. Use them for that.
Joel Webber retweeted
“The last remaining bottleneck is context on customers”
Well, it used to be before we shipped Subtext.
Give your Linear (or any) agent tools to efficiently “watch” every meaningful user session.
subtext.fullstory.com/docs/r…
first the execution is the bottleneck, then reviews (@linear diffs), now CI.
And the end the last remaining bottleneck is the context on customers, product and your understanding of it.
It's fun when they pick up on my jokes:
Me:
"Using `yaks` to shave its own yaks, should create more yaks, that help improve `yaks`. That's a great, big circular pile of yaks! A yak-ourobouros, if you will :)"
Agent:
"Notes I'm keeping for the ouroboros 🐂🔄: ..."
shopify.engineering/back-to-…
There's no longer any good reason not to write UI code in the platform's native language and core libraries.
I've spent way too much time building platform abstractions. There's no longer any point -- better to specify them clearly, and let agents do the grunt work.
Joel Webber retweeted
@indent Works great with @subtextdev out of the box.
It’s seamless to ask the Indent slack agent about customer bug reports or production incidents and automate the triage and root cause analysis with real user session data.
Introducing @indent, one coworker for your whole company
Indent is a single agent that is built to work with every member of your team, simultaneously. Instead of every one setting up their own agents, use Indent to build one together.
Try it at indent.com
Chip basically one-shotted this explainer with the interactive building metaphor. Other than a little z-fighting and a few other details, it looks pretty damned good.
Consider this grizzled old C hacker's mind officially blown.
Had to test out the Fable 5.1 hype for myself.
Verdict? LFG.
Check out this one-shot landing page concept that brings our new open-source project to life: sightmap.org/building/
Tl;dr: Few-shot working prototype of Cities: Skylines:
reddit.com/r/ClaudeCode/comm…
Frontier models have internalized so much of the techniques used in game engines, that "how do I achieve that" is ~solved.
Tired: What will humans do now?
Wired: What *new* things will we build now?
Joel Webber retweeted
Can someone help me understand whether iOS 27 MCP tooling will be compatible with WebMCP?
Here's how WebMCP differs from an MCP server for a website:
- MCP tools work even when the website isn't open
- WebMCP works on the page you currently have open. They use the page's current state and in your signed in session. Very helpful when you and your agent are collaborating.
For example, ChatGPT Work can use WebMCP tools in its built-in browser. So, if you're shopping on a site that has WebMCP enabled, the agent can work with the filtered search results, products, or cart that's already on the page.