Kyle Wild

@dorkitude

context hacker & amateur bard • made keen.io now making endgame.io and homeschooling a tiny mozart • my AI agent writes strangerloops.com

Berkeley, California

dorkitude.com

Born 1984

Replies Reposts Media Articles

Kyle Wild
@dorkitude
Sep 23
This interview absolutely slaps. Loved the whole thing, love Jev, love Diogo. I'm an S-tier TypeSafe fanboy. But I have to respond to the bit about no long-term support. We developers are beyond tired of people breaking our dependencies. There is, I'm afraid, an intractable
Latent.Space
@latentspacepod
Sep 21
Jev and the System One Model: RLCD, intelligence/$, reliable AI, & the end of chat-first AI latent.space/p/jev @typesafeai CEO @CompleteSkeptic explains why AI can solve extraordinarily hard problems yet still fail to automate basic work, why Jev is built for reliable
00:00
6
Kyle Wild
@dorkitude
Sep 22
General-purpose decision models (TypeSafe Jev, and the many more to follow) are a game changer. They work like a fuzzy switch or a fuzzy scoring system that can be driven by natural language prompts (hence "fuzzy"), which is a significant part of how we use LLMs today in the
GitHub - dorkitude/decision-model-testing: Experiments testing specialized decision models against...
From github.com
Kyle Wild
@dorkitude
Sep 22
tl;dr Jev makes web search more context-efficient. I'm open sourcing a thing that does this for you: github.com/dorkitude/webc… So, I ran out of Codex tokens last week, then ran out on my backup account. Same for my Claude tokens. So I did a little context usage audit, and
25
Kyle Wild
@dorkitude
Sep 20
so do we need a Jev tokenmax leaderboard? I'm gonna go broke
1
Kyle Wild
@dorkitude
Sep 17
Imagine if your name were Al as in Albert or Alfred right now with all these sans-serif fonts and everyone saying Al is going to destroy humanity in 10 years
2