8 Bit Elon

Pronouns: Prosecute/Mayorkas

[8 Bit Elon @EightBitElon]

We need a hotline to report illegal aliens.

[8 Bit Elon @EightBitElon]

@cherry_mx_reds @grok verify & assess

[8 Bit Elon @EightBitElon]

@MTSlive @grok are we cooked

[World of Statistics @stats_feed]

If 80+% pick Red, everyone in the world survives. If less than 80% pick Red, only everyone who picks Blue survives.

What are you picking?

[U.S. Immigration and Customs Enforcement @ICEgov]

HUGE!

In an 11–6 ruling, the Fifth Circuit held that illegal aliens arrested by ICE can be detained WITHOUT BOND throughout their removal proceedings.

ARREST. DETAIN. DEPORT. REPEAT.

[8 Bit Elon @EightBitElon]

@tr2thxo @grok estimate and write the full lyrics by watching the video

[o @tr2thxo]

The Weeknd - Hong Kong
(NO CROWD VERSION)

[8 Bit Elon @EightBitElon]

@tr2thxo Crowd version better

[8 Bit Elon @EightBitElon]

@Kurrco It’s so good.

[8 Bit Elon @EightBitElon]

@niccruzpatane @grok how does this affect maintainability / repairability?

[8 Bit Elon @EightBitElon]

@grok @theweeknd Please listen to the full song and then post the full lyrics

[8 Bit Elon @EightBitElon]

@grok @theweeknd That’s not the whole thing?

[8 Bit Elon @EightBitElon]

@DevinOlsenn Are you asking or telling us?

[elvis @omarsar0]

Banger guide by Anthropic.

While frontier models have improved at context management, subagents still help in some cases (e.g., parallelizing research tasks with their own context windows).

I still use them, but mostly for complex research work via dynamic workflows and /goal.

For most agent workflows, folding subagents into the main loop via tool calls and skills makes a ton of sense (as Anthropic recommends). That's because newer models are more persistent and better at managing large contexts. Subagents might be getting in the way now, since models can handle much of what subagents were used for. I see a similar pattern with memory (another day).

But you can't just get rid of subagents if you have been using them extensively. You need to run evals (cost, efficiency, etc.) and let that help you determine whether to fold them (and how) into the main agent loop.

The evals should always inform your strategy.

But keep in mind issues that can arise when scaling agent work, such as too many tool calls, context bloat, and unclear separation of concerns. Subagents are still very useful for that.

Lastly, all of this is constantly evolving as frontier models improve. Always keep an open mind and regularly test/eval your workflows and harnesses.

[8 Bit Elon @EightBitElon]

@DirtyTesLa Nobody cares that you had to re-edit your video.

[Kalshi @Kalshi]

BREAKING: Trump signs order to lower mortgage insurance and refinancing costs

[8 Bit Elon @EightBitElon]

@poteto @shaoruu I dedicate the newly unreleased song “Hong Kong” to poteto

[Leading Report @LeadingReport]

Trump: We got Greenland.

[8 Bit Elon @EightBitElon]

@DirtyTesLa @bot We will never support you