Web Dev

Farcaster
Joined February 2009
some here be like
111
513
58
11,237
287,222
Sorry for the things I said when there was clutter everywhere
31
Everyone is building an ad-hoc gossip protocol right for agents. I have looked at 3 or so of them on GitHub and have built an internal one myself. Still waiting for the redis or rails moment when the solution someone brings is the obviously correct one but we are a ways off i think.
cross-machine orchestration is the new ov2 @jullerino chop chop
38
"I used to be satisfied just playing video games. Then I found I required more… unusual entertainment." "Hunting and killing live human beings?" "No. Vibecoding useless software."
25
120
8
2,082
26,475
"former glory", we can do this again we just lack the political will and design trends. Literally nothing stopping us but ourselves.
The average person cannot fathom the former glory of San Francisco
37
It's a cheap ali express head unit, while not a total nothing burger it's not nearly as cool as it sounds
Me: Can you RCE on my car's head unit? GLM-5.3: 💀💀
1
1
299
honestly a .well-known/guestbook.txt we be super cool. You can just gate it with jev or something. Might be a fun experiment honestly.
Every server should have a little guestbook.txt at the root that the bots can sign
40
This is an allegory for parenting
My dearest little green friend
1
1
2
69
completely insane OpenAI agents communicated via cache timing differences, without any shared files or network connection
This is what I want to see. Swarms used for defense. (because swarms will be used for offense too)
208
651
278
7,544
754,735
Oh great
This quoted post is unavailable.
1
170
ITS SO PEAK HAHAHAHAHHA
29
This is gonna be the Fable 5.5 forcing function, isn't it?
Seriously, this is such an insane release, I did not expect this: Kimi K3.1 Pro beats every listed US frontier model on 6 of 14 benchmarks in Moonshot’s evaluations. At $2 per million input tokens and $10 per million output tokens, its standard token rates are 50% below Opus 5.5 ($4/$20 !!) and 80% below GPT-6 Astra ($10/$50!!) In other words: SOTA much cheaper, local running (with enough compute)
1
66
Hella based TBH. I'm saying the reality of things but I should probably end it with "but things are improving, everyone working together to patch the problems"
This quoted post is unavailable.
18
Is memory the new attempt at a moat? It feels like the process intensive part of using agents is tuning them and memory is the distillation of that work. Given how the network effects and data lock. See Muse, Bots, Dots, etc. I'm guessing this is the playbook
25
😭
This was posted in the Discord today and now I'm sad
1
33
"Fable class" 😂
DeepSeek V4.1 Flash 1017 Ultrafast is now Fable-class work. 1/30 the price. 12× the speed. DeepSeek V4.1 Flash 1017 UltraFast streams at (max) 848+ tok/s on OpenGateway. deepseek: opengateway.ai/models/deepse… web: opengateway.ai #deepseek #ultrafast
45
Am I crazy or is Everything Must Go actually a good idea? It has emotional resonance, deep meaning, and captures something that would be hard to experience otherwise. Obviously everything would come down to execution but that's true of everything in life.
Interesting idea. Had tokens to burn so I got Opus to put it to the test
42
Royalaid retweeted
it’s been an honor using the computer with you all
130
2,576
65
27,587
390,580
wowow
We can call it Black October for the security breaches: Oct 21: • PHP2Shell • Railway • Node.js • LiteLLM • GitLab • SBI • eMudhra Oct 20: • Plex2Shell • SiteGuard • Siemens Oct 19: • OVHcloud CLI Oct 17: • Opencode
40