Eugene Oldman

Founding engineer. Building agent harnesses and dev tools

[Eugene Oldman @_cmd8]

https://t.co/zxGNoucLSS

[Eugene Oldman @_cmd8]

@Jason bs

[Eugene Oldman @_cmd8]

@digitalshane_ yup, we are def cooked, not even "medium", "well done" all the way

[Eugene Oldman @_cmd8]

@barre_of_lube :(

[Eugene Oldman @_cmd8]

@barre_of_lube is there a cli or mcp to drive it with agents?

[Eugene Oldman @_cmd8]

@Voxyz_ai liked this one too https://t.co/QuXFYPl5AP

[Zihua @luinlee]

最近用 Opus 和 Astra 写了不少 UI，发现一个避免 AI slop 并且让界面更好看的秘诀：

在 AGENTS.md 里加上 “Avoid unnecessary secondary labels and descriptions. Let the layout do the explaining.”

[Eugene Oldman @_cmd8]

@charlielamb stopped using arc ages ago, zen has been good enough, helium is catching up

[Eugene Oldman @_cmd8]

@barre_of_lube that's all I need to know, ty

[Eugene Oldman @_cmd8]

@barre_of_lube does it support ublock origin (not lite)?

[Eugene Oldman @_cmd8]

Cyber Wednesday

[ThePrimeagen @ThePrimeagen]

Hey, it is Wednesday

OpenCode hacked
eMudra had picked a private key value of 4, thus everything is now deemed unsafe.
PHP, running 70% of the internet, has an insane hack...

Happy getting internet getting humped day

[Eugene Oldman @_cmd8]

@EXM7777 probably using jev for routing

[Eugene Oldman @_cmd8]

@dotpem love jev https://t.co/8TqefJCBR0

[Eugene Oldman @_cmd8]

https://t.co/zxGNoucLSS

[Eugene Oldman @_cmd8]

@WatcherGuru

[Eugene Oldman @_cmd8]

@mxstbr are we reinventing webhooks?

[Eugene Oldman @_cmd8]

@90_mp3 @Ibelick that's literally what storybook is for, add visual regression to it and interaction testing and you're on the right track to reduce slop in the codebase

[Eugene Oldman @_cmd8]

jev++

[Neal Patel @neal_k_patel]

You already gave the LLM the answer. So why is it writing it out from scratch?

Introducing a new AI model architecture: Constrained Decoding Models (CDMs)

LLMs were built to write. Most production workloads just need an answer: fast, consistent, and affordable.

CDMs are purpose-built for 1 task each. They read your input once and skip the token-by-token generative loop that makes LLMs slow and expensive.

Not just a decision model, but a family of 3 CDMs live today:

• Classify: define your labels and get back a calibrated probability + reason for each. Used for routing, tagging, and intent detection.
• Extract: describe the entities in natural language and get back well-formed JSON with confidence scores + citations. Used for web scraping, document processing, and PII redaction.
• Summarize: pass in long text, images, or PDFs and get back a consistent summary under 1,000 tokens. Used for context compaction, metadata generation, and transcript digests.

Built for production:

• 1M tokens of multimodal (text + image) input, 31x larger than Jev
• 77.2% on DecisionBench vs Jev's 72%
• 200x faster, 467x cheaper than frontier LLM APIs
• $0.05/M input tokens. Output is free (yes, even for summarize)

Live on your favorite platforms today:
@OpenRouter @vercel_dev @LiteLLM @LangChain @TrustedRouter @itsArthurAI @useRouterPlus

Get 50M input tokens for free (for the first 50 hours) at link below 👀

[Eugene Oldman @_cmd8]

@beffjezos @Grimezsz absolutely NOTHING can go wrong with that!

[Eugene Oldman @_cmd8]

@jonnymevs email + password (in a password manager), everything else is a degraded UX