Peter Schuller (@scode) profile picture

Human. I'm KZY2KZZH on Threema, scode.01 on Signal.

Los Angeles
Joined February 2009
Photo by Peter Schuller (@scode)Photo by Peter Schuller (@scode)Photo by Peter Schuller (@scode)Photo by Peter Schuller (@scode)Photo by Peter Schuller (@scode)Photo by Peter Schuller (@scode)Photo by Peter Schuller (@scode)Photo by Peter Schuller (@scode)Photo by Peter Schuller (@scode)Photo by Peter Schuller (@scode)
Found bugs in a couple of upstream projects that ban the use of AI not only in actual code submissions but in bug reports describing the bug. Result? I'm simply not going to report them. I considered starting to just dump bug reports somewhere where these people have no right to create rules so that at least it can be found by others or their agents. The notion that I'm going to manually write something up because someone is hysterical about AI is ridiculous. Demand a quality bar sure, stop being opinionated about tools. Imagine banning code written on a computer because all submissions must be written using punch cards. That's how ridiculous this is.
4
107
Tokens are burning ripping out all the web stuff from farhelm.io/ (replacing with GPUI in preparation for Linux desktop support). It should be interesting - also baking in code improvements prioritized based on analysis of the past few months of bug fixes due to agents not applying reasonable good practices. Woke up to screenshots of an early milestone this morning: snippets.scode.org/s/farhelm…
1
1
118
Also if a rule or regulation seems to cause excessive complexity or friction, escalate to re-assess weather it's worth it. Do not proceed until fable, astra, sol and a fresh opus consult all agree it's worth it.
claude, rewrite the FAA in Rust, make no mistakes
2
190
s/fire/allow/ and this is me yelling at macOS to stop getting in the way of my agents with privacy/accessibility protections in VMs...
1
1
86
I so so so hope this is successful and does what it says on the tin (and/or others like it). This needs to be super cheap for both sides and just the default for everyone all the time. openai.com/index/introducing…
73
I stand corrected even worse.
1
109
The most ridiculous "safeguard" block I have ever seen. Literally just having a design/goal planning session for a major change and just wanted to record a verbatim copy of our planning conversation. Apparently it's triggering a "reasoning_extraction" safeguard.
2
92
I saw someone today say they couldn't imagine what could be improved upon relative to Opus 5.5. Today I was talking to an agent about an unattended plan that had gotten bounced back and forth between agents a few times due to complexity spirals, and now got blocked again (this time just Opus involved, no Sol). I'm like why is this so complicated. It gives a long winded answer involving solving a distributed systems problem involving deleting a Tensorlake sandbox if and only if a certain agent session is the *only* session running on it (managed by a supervisor process). So... Agent: <long winded explanation of why it's so hard after it bounced between agents N times> Me: i don't get it. what is so stupidly hard about just having a protocol message that goes to the supervisor that says "kill this session and disallow all future sessions" which is atomically either accepted by the supervisor or not? (it only succeeds if it is the last session) Agent: Nothing about that is hard, and it's the right design....
1
1
111
This is happening repeatedly nowadays. I'll find that sol is still working on something after many hours. I'll ask an agent to troll the transcripts and tell me if it's going off the rails, and I found that yeah something went off the rails into needless complexity. Then I have to intervene. I think I'm going to start having an agent monitor transcripts and start automatically nudging instead of waiting for me to wake up in the morning.
1
2
236
I feel the importance of models actually being able to communicate sensibly to the human is growing increasingly important. The more I rely on models at a much higher level rather than to produce code I read line by line, the more important it is to be able to speak in a language and with concepts that make sense - at the right level of abstraction, and w/o involving irrelevant implementation details. It's funny because what models are doing is very similar to what real software engineers do all the time. I'm sure everyone who's a software person has heard a software engineer give some presentation in front of 10 people and it's just obvious no one in the audience has an idea what they're talking about, or you've gotten help by another team in your org and they speak in internal jargon instead of customer visible product surface terms. Models just are very extreme at doing it. I like to think of it is - they are extreme experts at whatever sits in their context at the time and speak as if the reader is too. Which is basically exactly what humans do. :)
1
1
155
Another day, another arbitrary Apple imposed limit I learn of. Apparently Apple has decided I can't run more than two macOS VMs on my mac mini, that I specifically got because Apple has decided no one gets to run their os on non-apple hardware nor dole out compute time in chunks less than a day.
3
6
1,133
I understand making faster playback beyond a threshold a "premium" option like youtube. But if you're not upselling (in this case Patreon), why don't you just allow faster than 2x? It's another option. So cheap to just make the limit 3x or 4x instead of 2x.
118
WhisperX + AI agent is doing a really good job editing audio voice over for me. It appears this will trivially remove all the tedium of editing away pauses, stumbles and re-takes. I can't believe I didn't set this up before. All on a remote Linux box. The screenshot is of a throw-away web interface using Remotion Studio. Took me less time to set up than it's taken me to edit even a single voice over narration in any one past video.
1
72
Surprised so many model/harness combos have trouble waiting for a scripted event. claude code seems decent at it by default. Codex sucks - to the point that I have a whole skill just with a pre-written script and harsh instructions for the agent to not burn cached input tokens polling all the time (github.com/scode/dotfiles/tr…). Grok bot also just randomly failed to succeed in having a script wake itself up.
2
227
Apparently I need to adjust my skills or claude code will refuse to write sub agent reports to files even when explicitly told to.
2
3
450
If you've been hearing about the Starlink/Ukraine thing over the past week, you might be under the impression that what happened is that they slowed down internet speeds. Well, that's wrong. I only realized because I accidentally stumbled upon reporting that wasn't misleading, after only hearing misleading reporting for days. I had a bunch of agents work on a TLDR of what actually happened and how it's been reported: snippets.scode.org/s/starlin… (I am making a point of quality of reporting, this is not an opinion on anything else.)
1
1
127
Well, this is highly unscientific but better than "I used it for a while", and definitely not benchmaxxed: snippets.scode.org/s/rust-co… Quality variance is basically noise at this sample size but still consistent with same ballpark performance - and way faster than both Luna and Haiku.
1/4 Today, we’re releasing Mercury 3. Mercury 3 is by far our most powerful model and the most intelligent diffusion LLM ever released. It outperforms GPT-6 Luna and Haiku 5.5 on quality while running at 1000+ tok/sec.
104
codex and claude code are so yesterday, let me introduce you to Palantír
1
2
132
I just tried it, this is potentially exciting! I have only done basic stuff (repo navigation, listing PRs), but 2.5 fell immediately flat when I tried it. This worked fine, and it's *so so so* fast.
1/4 Today, we’re releasing Mercury 3. Mercury 3 is by far our most powerful model and the most intelligent diffusion LLM ever released. It outperforms GPT-6 Luna and Haiku 5.5 on quality while running at 1000+ tok/sec.
2
11
578
Haha. I was daily driving a claude/codex load balancer I am building, and a bug caused a session to be unrecoverable. The other session which was debugging the problem instead hit a bug in codex-lb so now *that* session is unrecoverable. (Luckily just asking an agent to read the other agent's transcript more or less works pretty well as a fallback resume mechanism.)
4
1
268