Xponder

is this real?

[Xponder @expondR]

In the future, if we get to AGI, I think the world will be forced to introspect. Everyone will have to go inside their void. The whole human race might become curious and not just a few individuals.

If it's sudden, it will be devastating. But I think we will ease into it. The internet has already amplified the dimensions of our collective thinking. AI is definitely gonna expand it even more.

https://t.co/vemI8ZAaVx

[Xponder @expondR]

@programmergrind Lets go!!!

[François Chollet @fchollet]

The ideal outcome for science isn't a couple of labs using proprietary models and millions of dollars to crack trophy problems, it's to have millions of AI-aided human scientists who follow their curiosity and become way more ambitious

[programmergrind @programmergrind]

claude surprises me too

[Xponder @expondR]

It's the same with artists. Now there are tools that can generate art as good as them however they fail to realise that it is all about the end product and not what tools you used to build it. Ego has no place while striving for excellence.

[DHH @dhh]

It was almost impossible for a programmer who wrote all their code by hand to divorce their ego from the output. Now agents are delivering what decades of agile admonishing about “collective ownership” never could. What a gift!

[Xponder @expondR]

Omarchy

[Xponder @expondR]

Only philosophy can save you from suffering. A philosophy that lets you know about your place in this universe.

[Xponder @expondR]

For some questions in life you just have to wait and let the answer show itself. There is no speed running understanding and realisations.

[Xponder @expondR]

Tools are never the bottleneck. Imagination is.

[Xponder @expondR]

Animated lockscreen in omarchy.

[Xponder @expondR]

For some reason astra is lightyears ahead in quickshell compared to any claude models. It is honestly mind boggling. This is the first time in a long time i have been having fun with an openai subscription.

[DHH @dhh]

With the latest results from the @rails agent evals, it's clear to see that @openai still has a solid lead, despite Opus 5.5 making a good jump. The dark horse here remains Luna Max. 18% completion at just $11! Not that far off GPT-6 Sol!

[Xponder @expondR]

It's funny how the experience is completely opposite for someone who has no programming background. There is not much to do except verify the output. Reading code is not really an option. Verifying is easy for frontend applications. Backend is where it gets tricky.

[François Chollet @fchollet]

Delegate coding. Never delegate understanding.

If code was previously your source of truth for managing understanding, then you need a new source of truth artifact. If code-centric workflows were how you developed that understanding, then you need new workflows.

[Xponder @expondR]

@awkwardgoogle @grok what's the amount of g-force she experienced? Approximately.

[Xponder @expondR]

@ojasaklecha @arpit_bhayani Ofcourse. Your main agent will miss a lot of things too. The good thing about subagents is you're not the one prompting it.

[Xponder @expondR]

@ojasaklecha @arpit_bhayani That's the thing, if the subagent follows the instructions properly, why do you need its thought process? Your main agent prompted it so you'd expect it to do what it was told. Your main agent should verify this by default.

[Xponder @expondR]

@arpit_bhayani The other two problems you mentioned sounds pretty valid. But i think the main agent should verify everything anyway and then you review the whole diff but it is still definitely a problem without a clear way to trace the source.

[Xponder @expondR]

@arpit_bhayani Isn't the main agent giving them fresh tasks where the context between you and main agent doesn't matter much to the subagents?

[Xponder @expondR]

@mark_k Would love to be a part of this community.

[Xponder @expondR]

The things you can do with quickshell in omarchy.