←
Neil Roberts10,542 posts
Neil Roberts avatar

Neil Roberts

@pottedmeat

🇧🇸🇺🇸 TypeScripter @SitePen (formerly at @Automattic), LLM enthusiast, Tabletop playtester, 👍 Theme parks, Sports Racer, Nerdfighter, 8w9

◷ Joined August 2007▧ 293 Photos and videos
Nick Nisi avatar
⟳ Neil Roberts retweeted
Nick Nisi@nicknisi·Sep 3

I made a /brainrot skill to meet you where your AI smooth brain is at and teach you stuff.

1:38
Neil Roberts avatar
Neil Roberts@pottedmeat·Jul 28

If you're a visual designer, the right approach is definitely to have as many unhinged unsolicited semi-accusatory rants about AI as possible with clients to make it perfectly clear how AI offers no advantages over working with you.

Neil Roberts avatar
Neil Roberts@pottedmeat·Jun 5

I do enjoy that some of the worst software being made right now is the AI coding tools themselves.

Nick Nisi avatar
⟳ Neil Roberts retweeted
Nick Nisi@nicknisi·May 9

"Where was that conversation where I figured out the auth bug?" If you use Claude Code, Codex, or Pi across multiple projects, you've lost a session. It's buried in a JSONL file somewhere in ~/.claude/projects/ and good luck finding it. So I built `sessions` One command to browse every Claude Code, Codex, and Pi conversation you've ever had. Fuzzy search what you typed, filter by tool, scope to a repo. And if you use git worktrees, those sessions are basically orphaned the moment the worktree is cleaned up. `sessions` still finds them, flags the directory as gone, and tells you exactly how to resurrect it. brew install nicknisi/formulae/sessions

Post attachment
José Valim avatar
⟳ Neil Roberts retweeted
José Valim@josevalim·Apr 27

At this point, this is just irresponsible. Yes, coding agents are leading to an increase of software production, but we are not seeing a similar push or increase in software quality. If Anthropic focuses on safety and it believes software engineering is going away, then it needs to be doing much more to improve how we design, build, test, and maintain software (aka software engineering). Increasing the production of unreliable, poorly designed, and unverified software directly undermines safety. Claude Code is claimed to be "fully written by AI". In the last two months, it took three separate postmortem-worthy failures and user complaints to surface what their own testing missed. Yesterday users were being over billed by hundreds of dollars. Software engineering isn't ready to go away and there is not enough progress to argue that case. I am certain Anthropic would argue that AI progress in other domains is strongly dependent on having proper safeguards in place. I can't wrap my head around the cognitive dissonance when it comes to software. PS: Mythos (may) improve software security, but that is only a subset of safety.

Anthropic CEO (Dario Amodei): "Coding is going away first, then all of software engineering." What do you think about this?

8:46
Speculator avatar
⟳ Neil Roberts retweeted
Speculator@TheSpeculator0·Apr 16

Working at Anthropic must be like being on crack. Get paid a million bucks a year to --dangerously-skip-permissions vibe your way to releasing a new product every day. Does it work? not really. Is it reliable? also no. It doesn't matter, you're building the machine god.

I feel bad dunking on them so much but it's genuinely absurd how bad the new Claude Code desktop app is. You can feel the vibe code leaking everywhere. Every "feature" is barely integrated and full of edge cases that weren't considered. Every menu feels barren, stuffed in last second for some random toggle. Every hotkey breaks as soon as you try to do anything else. I've lost track of how many bugs I've encountered. I found at least 40 in under an hour. And it's all truly absurd arcane shit. Stuff like voice mode typing in all input boxes instead of just the one you have focused. Any one of these issues would have been enough for me to do a massive post-mortem and likely fire someone. A $400b company shipping this is absurd. I feel like I'm going mad. How does anyone seriously use this?? It is broken on fundamental levels that are hard to comprehend. How are we supposed to trust the code these models produce if Anthropic's official showcases are absolute slop? Dedicated video on this coming tomorrow. Just needed to get this off my chest.

0:07
Convex avatar
⟳ Neil Roberts retweeted
Convex@convex·Apr 8

convex 2.0 has been online internally since february. when first activated, it synced data that didn't even exist yet in response, we've made a "convex lens" program with select partners. a think tank to prepare the world for this level of DX if you're not scared, you're ngmi

Neil Roberts avatar
Neil Roberts@pottedmeat·Apr 2

Seeing someone curled up asleep on a park bench: “Why don’t they get a job?” “They’re probably on drugs.” “This is what’s wrong with the city.” Seeing someone curled up asleep on a bench in the airport: “Poor thing, they’ve had a rough night.”

Neil Roberts avatar
Neil Roberts@pottedmeat·Mar 20

Our original wait time was 50 minutes. We’ve literally been stuck in the same spot for 15 minutes. And this is the second time we’ve had an extended sit.

At @UniversalORL the way they implement their express pass, you can join a line with one quoted time and end up waiting significantly longer as they prioritize pay-to-play. In the line I’m in, everyone has just given up and is sitting in the floor.

Neil Roberts avatar
Neil Roberts@pottedmeat·Mar 20

At @UniversalORL the way they implement their express pass, you can join a line with one quoted time and end up waiting significantly longer as they prioritize pay-to-play. In the line I’m in, everyone has just given up and is sitting in the floor.

Josh Daws avatar
⟳ Neil Roberts retweeted
Josh Daws@JoshDaws·Mar 19

I've been reading most books on Kindle for 15 years. Now my digital assistant has access to all 11,328 highlights across 15 years and hundreds of books. This is what I mean when I talk about giving AI the right tools. "Give me everything I've highlighted about the church growth movement from David Wells."

Introducing the Readwise CLI. Anything you've saved in Readwise (highlights, articles, PDFs, books, youtube, newsletters) is now instantly accessible from the terminal. For you, and your AI agents. npm install -g @readwise/cli

1:18