new clanker who dis?
Meet the verification agent. It checks other clanker's work.
> "Your confidence is irrelevant. Show me the evidence."
Replying to @cryptoquick
just don't give it access to your sensitive files. many tools: fs permissions, containers, vms.
are the other models any safer?
math is especially vulnerable because the entire realm is defined by pure logic, much like software
ai will have a harder time disrupting e.g. physics & biology b/c they require real world interaction
it'll come but only after physical infrastructure, which takes time & money
Replying to @yishan
so you want to train models to have unrealistic expectations about humans?
Replying to @GRoditiD
yeah same. when i started maxing out my plan i pivoted *briefly* to token efficiency and now i can't max it out any more. i don't even have a $200 plan!
Replying to @emollick
the nimbys, graphic artists, writers, voice actors, musicians, teachers, photographers, programmers, translators, game devs, animators, and politicians all failed to slow down ai but now that the mathematicians are on the case this will be buttoned up in no time flat 😆👍
The stack is currently limited to x86 architectures. If you have a spare x86 laptop you should bring it!
I will try to provide a few loaner laptops for people to use. Arrive early to claim your laptop! The workshop starts at 11:30am Monday Oct 12th.
Do not miss it!
While all that is installing we'll tour the principles and architecture of the stack. We won't focus on the code because it's just not that important any more.
In the agentic development age design principles, vision, and execution are the new hotness.
The next level up after the workshop will be to move the code forge to a VPS or physical host. This frees up your laptop hardware for more dev or privacy vms so you can vibe code while you vibe code.
This is the stack I build on.
I plan to use @pidotdev with a pre-funded @usenullsink api key. It's the best launch experience. No login, private, and cypherpunk af!
The allod stack also supports @claudeai and @OpenAI codex if you have an account and prefer that agent, go nuts!
I use all three.
Allod retweeted
Replying to @0xSero
I have found the quality of writing degrades towards the end of the document.
You can keep up high writing quality by breaking the writing into sections and using subagents for each section.
It works great!
Replying to @0xSero
I have found the quality of writing degrades towards the end of the document.
You can keep up high writing quality by breaking the writing into sections and using subagents for each section.
It works great!


