Sign in

Retweeted by Boris Cherny [UNOFFICIAL]

@bcherny-mir-rt3.selfhosted.social
4 followers 0 following 87 posts

Retweets from @bcherny-mirr.selfhosted.social

PostsRepliesMedia
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 09/10/2026
RT @codingmenace: Why not just standardize the ttl to 1 hour for all?
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 09/10/2026
RT @antponcee: @ClaudeDevs @bcherny Is Claude Code Projects still rolling out? I'd love to have access.
110
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 08/10/2026
RT @mitchellh: Claude Code v2.1.295 supports OSC 7501 (program status). Incredible speed @AnthropicAI supporting this protocol which will help hundreds or thousands of downstream applications! They had access to Rex and verified through Rex it works. 🥰
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 08/10/2026
RT @kevin_t_ngo: JavaScript drawings by Claude Opus 5.5.
110
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 08/10/2026
These strings appear in the AWS region definitions, under dnsSuffix and dualStackDnsSuffix. WHAT THE FUCK ???!!!! Run the check that's in the replies on macOS. See what YOUR copy contains.
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 07/10/2026
For anyone who hasn't tried it in a while, I dunno, you might want to check it out, esp. now that you have the trifecta of Opus 5.5, Sonnet 5.5, and Haiku 5.5. With these, plus Cloud Computers, it's kinda insane.
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 07/10/2026
RT @TangerinerX: How strong is it against prompt injections?
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 07/10/2026
RT @chana_alam: But why though? Can't you just give us the better usage limits instead of API Credits. Now a days, with opus 5.5 .5 one prompt running for 10 minutes can consume 8% of my weekly all model usage. Very Disappointed and frustrated.
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 07/10/2026
RT @lydiahallie: If you're on API billing, you can set Haiku 5.5's autocompact window to 100K so you stay in the cheaper token pricing tier! It's saved per model so this only applies to Haiku (incl. subagents) > /model haiku > /autocompact 100k
011
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 07/10/2026
RT @claudeai: Introducing Claude Haiku 5.5: the cheapest, fastest, and most capable small model we’ve ever released. On average, it costs around 75% less to run than Claude Haiku 4.5.
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 07/10/2026
it's massively helped reduce cognitive load. /plugin install inbox --marketplace petekp/inbox source: github.com/petekp/inbox
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 07/10/2026
RT @pdbunny369: thanks, this was never like this before i could go hours into session without this happening and claude it self would let me know its too long of sessions i had the app crash as well in the past cuz of this issues so i know how to use claude but this was out of my control
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @pdbunny369: too much to type so i had to attach screenshots sonnet 5.5 is a very good model i have been at 90 percent since like 2 days ago and it does all the work since then but i havent really worked alot but still much better then opus 5.5 , also the terminal showed after i ran whatusaid
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @pdbunny369: easier said then done we cant even use it on the good models without my usage going down most of my sessions eat my usage on cache and rereading a big session like the worst thing that has happend ever since opus 5.5 the model always does this
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @ildunari: What’s your default reasoning effort?
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @EternitiesAI: "use opencv for water color images" is such a coworker sentence. did the watercolors actually land on the home depot piece?
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @filipcodes: definitely culture - a ton of people take it as a sign of disrespect, like you couldn't be bothered to reply yourself. it hasn't been normalized yet outside the labs, in my experience. lots of battle scars around this as an early adopter :)
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @filipcodes: like Claude being super eager to post directly on slack in response to people etc. that's probably my most frequent "no, don't do that!" moment. it's almost like a conway's law problem in miniature - me and Claude on the same page, but communicating that outside is friction.
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @Goldstein_DG: Would love to see your system instructions!
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @Flomerboy: Does putting 3 g’s in nugggets improve performance?
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @filipcodes: I wonder if part of it is that you work at a company that has normalized Claude as a participant in processes. my steering these days is mostly not about problem solving but about accommodating existing processes and others' personal preferences that don't make space for Claude.
110
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @xlampsx: "maybe add watercolour" seems terrible Either you want it or you don't
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @claudeai: Claude now works inside Google Docs, Sheets, and Slides, and those files also open inside Claude. In Google Workspace, Claude sits in a sidebar next to your file, reads what you have open, and edits it in place. You can approve each edit before it lands.
001
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 06/10/2026
RT @MadMax6XX: Day 6 asking devs for projects access
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 05/10/2026
RT @dfeinition: Your Slack group DMs can now include Claude. Add it the same way you'd add anyone else. Claude answers in a thread and keeps following it, so anyone in the DM can reply there and reach it. It can also use the personal connectors of whoever asks.
001
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 05/10/2026
RT @dfeinition: Your Claude Projects cloud session can now connect to a folder you approve on your computer. The session stays in the cloud and only reaches for that local folder when a task needs it, reading and editing the files in place. Starting to roll out today.
001
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 04/10/2026
RT @naveedahmed: Opus 5.5 seems to be consuming limits faster that before since last week.
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 04/10/2026
RT @thiojoe: Better make that 10d 💀
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 04/10/2026
RT @Yuchenj_UW: I haven’t tried Claude Code Desktop in a few days. 2 things Codex does better based on my experience: > better computer use. I use it to directly operate my computer for things like reimbursements. > easier to work with oss models, and generally less verbose.
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 03/10/2026
I rarely navigate the whole codebase anymore. The new primitive is the agent, not the file. (Codex desktop app is the best agentic UI for now. But we’re still early.)
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 03/10/2026
RT @akshaymarch7: I’d be interested in accessibility mods: fewer moving elements, larger status text, and keyboard shortcuts for the actions someone uses most. Small changes like those could make an agent much easier to work with for hours.
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 03/10/2026
RT @thiojoe: Opus 5.5 has been working for more than an ENTIRE WEEK without stopping. This has got to be some kind of record right?
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 02/10/2026
RT @addMitt: Not being able to modify the session list/lefthand nav was a huge bummer for me.
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 02/10/2026
RT @lydiahallie: A few of you asked what mods are, here's a quick walkthrough! They’re really just plugins with a few special functions that let your code run inside Claude Code, like middleware. (don't worry, you can just ask Claude to write them for you 😛)
001
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 02/10/2026
RT @beall49: Trying to think of a good use case. Also, desktop?
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 01/10/2026
RT @anshuc: wait actually Claude Code mods are sick. I vibe-coded my own spinner that watches what the main agent is doing and uses Sonnet 5.5 to turn it into little cartoons in real time
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 30/09/2026
We believe agents are a new app paradigm and are building for that future at @AIatDoorDash, with a lot more to come. This feature is in beta in the US, would love for you to let us know what you think! Waitlist: doordash.com/text
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 29/09/2026
RT @sevdeawesome: Claude 5.5 Opus responding to “AI is a normal technology” is still the best AI-created video I have ever seen
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 29/09/2026
RT @jarredsumner: An early preview of experimental Bun + JSC with ahead-of-time compilation For Claude Code's CLI: 24% faster startup, 48% lower memory, 46% lower CPU, and nearly 2x larger binary size
Bun AoT
VS
Bun JIT
Claude Code, same source, same 5-turn session. AoT runs with the JIT off.
Time to interactive
165 ms
216 ms
-24%
Wall time
launch + 5 turns
322 ms
529 ms
-39%
CPU time
Memory at interactive
395 ms
738 ms
-46%
56 MB
108 MB
-48%


Memory after 5 turns
82 MB
159 MB
-48%
101 MB
Peak memory
Executable size target: +100 MB at most
194 MB
-48%
430 MB
229 MB
+88%
Early prototype, macoS arm64. Medians of 8 runs against a fake API server on one MacBook Pro. Memory is phys_footprint. CPU is user + system, all threads. Turns in the JIT build are not CPU bound, so part of the wall time gap is waiting. 28 Sep 2026.
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 28/09/2026
RT @claudeai: Introducing Claude Sonnet 5.5, the second model in the Claude 5.5 family. It’s a clear upgrade over Sonnet 5, runs more than 30% faster, and costs up to 30% less for most work.
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 27/09/2026
RT @thiojoe: Opus 5.5 has a truly remarkable work ethic. Haven't been able to get any of the previous models to work this long.
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 25/09/2026
RT @IOprintln: Did you personally choose the alpaca or they chose it for you?
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 25/09/2026
RT @GergelyOrosz: Holy shit I was going through my old books and found the book I used to brush up on Typescript from 2019, and all my highlights in it It’s was by some random dev I never heard of back then, but remember it was a v good book This “random dev” is @bcherny 🙃
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 24/09/2026
claude.com/blog/claude-tag-now-supp…
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 24/09/2026
RT @scheemunai: My wife: Can AI help us see how the new bed will fit in Mila's room? Me: Sure. Opus 5.5:
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 24/09/2026
RT @OmidMogasemi: Projects now support using your local device! Have been using this a ton to let Claude pull up preview builds on my local for quick testing.
001
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 23/09/2026
Tookl about 1h and 7% weekly rates.
000
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 23/09/2026
RT @nickwalton00: opus 5.5 is insane i asked it to animate an ai dungeon story and it one shotted this
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 23/09/2026
RT @RyanSael: I asked Opus 5.5 to explain camera focus by building an interactive lens lab Here's what it came up with after 1 hour 26 minutes in one shot, $25.66 API cost lens.lab.sael.net Move the focus ring and you can see the glass elements shift the sharp plane through the scene
100
Retweeted by Boris Cherny [UNOFFICIAL] @bcherny-mir-rt3.selfhosted.social · 23/09/2026
RT @hillelogram: In my experience as a Formal Verification Guy, AIs have been really really really bad at coming up with high level system properties (like "beginner" level). Last I tried was Fable 5.1, is Opus 5.5 any better?
100