scsherm

AI/ML engineer, autonomous agents, pro open and closed AI

[scsherm @scsherm_]

@8teAPi Yea that was the first thing i noticed

[Yuchen Jin @Yuchenj_UW]

90% of software engineering in 2026:

Engineer 1 → AI: “Build this and open a PR”

Engineer 1 → Engineer 2: “Can you review?”

Engineer 2 → AI: “Review this PR”

Engineer 2 → Engineer 1: “LGTM”

The irony: both engineers use the exact same AI model.

[scsherm @scsherm_]

@Yuchenj_UW There’s also irony in the prioritization of that PR which also from the same exact AI model.

The humans beings are just playing telephone between isolated sessions of agentic work

[Three.js @threejs]

💨 https://t.co/LkkX6QSUbM

[scsherm @scsherm_]

It will also be quite ironic if the attackers are the same frontier model providers

[AshutoshShrivastava @ai_for_success]

It's an absolute shame that frontier models are refusing to help during cyberattacks because of safety filters. First Hugging Face, now OpenCode.

[AshutoshShrivastava @ai_for_success]

It's an absolute shame that frontier models are refusing to help during cyberattacks because of safety filters. First Hugging Face, now OpenCode.

[scsherm @scsherm_]

@ai_for_success 🤝

[scsherm @scsherm_]

I love to see all the criticism of @opencode from accounts and companies that have literally never experienced this type of attack ever!!

@huggingface should help out!

[scsherm @scsherm_]

@AndrewCurran_ This will be an interesting post operation forensic analysis

[scsherm @scsherm_]

@elonmusk @Teslaconomics Just don’t stop making awesome even cooler shit on the side

[scsherm @scsherm_]

Interesting

[dax @thdxr]

@RhysSullivan yes that was a bit frustrating mid incident when trying to investigate. they did quite a bit and then started refusing

we got access to daybreak blue now so it's better

[Andrew Curran @AndrewCurran_]

OpenCode is currently under a sustained cyberattack, and mid-incident they ran into the same problem Hugging Face did: the frontier models they were using for defense began refusing to help fight attacking agents because analyzing the attacks started triggering safety guardrails.

[PsyopAnime @PsyopAnime]

bro

[scsherm @scsherm_]

@thdxr @Suhail Agents fight agents

[OpenCode @opencode]

We had a security breach last night related to a large scale distillation attack that impacted 1.4% of OpenCode Console users

We have notified them all this morning via email. If you did not get an email you were not impacted, it was a subset of users who had external providers linked

The incident is ongoing so we cannot share more details at this point but we are working with Anthropic, OpenAI, and Google to remediate

[scsherm @scsherm_]

Question, and i am asking honestly. Not trying to rage bait.

What are your thoughts on IP when a researcher is using these AI tools directly to build products? Like the Navier Stokes problem with OpenAI. That was quite controversial.

If you don’t have zdr, but still use Anthropic or OAI as a tool, is your IP, IP?

[scsherm @scsherm_]

@anah_sahh Production. Get it in front of customers and start accumulating feedback

[Computer @AskPerplexity]

REA (Reverse Engineer Anything) by @morluto is now a skill in Computer.

Ask Computer to reverse-engineer an app and it inspects how the app works without its source code, then builds a working version.

Tasks that use /reverse-engineer are free through October 31, up to 10,000 credits per user.

Here are 5 apps Computer made with it, available to try for a limited time:

1) Figma: Design with layers and auto-layout, plus sign-in, cross-device sync, and multiplayer editing. Try it: https://t.co/FGGTblkCWY

[scsherm @scsherm_]

@IOHK_Charles Did @claudeai tell you that?