danfiru

Life turned into a game of Factorio: automate everything.

I’ve been rolling with 2 for a while. 5 sounds even better. Especially for work with more subjective success criteria at the outset.

@Hesamation Perfect use of the meme.

claude code users waiting for the inevitable price bump on the subscription they atrophied their entire coding skills for.

@dkundel I’m gonna use the hell outta this until it eventually gets shut down.

I built a new plugin! You can now trigger Codex from Claude Code! Use the Codex plugin for Claude Code to delegate tasks to Codex or have Codex review your changes using your ChatGPT subscription. Start by installing the plugin: github.com/openai/codex-p…

@Teknium @WilliamLamkin LFG

It's FINALLY HERE! Multi Agent Profiles so you can have as many independent bots with their own memory, gateway connections, skills, chat history, everything! To use: Run `hermes update` and look for multi agent profiles User Guide: hermes-agent.nousresearch.com/docs/user-guid… Reference: hermes-agent.nousresearch.com/docs/reference…

Apparently you can just add “what would John Carmack do?” to your prompt and you get way better results.

Whoever in Japan looked at an orange and said ‘nah, I can do better’ and then created the Sumo deserves to be carried through the streets.

At this point i'm pretty sure claude code is RLHF'd into telling you not to switch no matter the objective criteria. I asked it to evaluate switching a subagent to Hermes Agent and it keeps saying "NO"

The weekend tokens hit different.

I'm building a 200 person company that punches like a 20,000 person company. Watch.

Yesterday I used “trad code” in a sentence and no one batted an eyelash.

NPU poor is the new GPU poor.

@Teknium Nooooo you’re gonna get one shot. We have work to do!

Look what arrived last night ^_^

@ziwenxu_ Hermes Agent

As a founder, What do you prefer? 1. OpenClaw 2. Hemers Agent 3. Claude Cowork

The night time tokens just hit different.

my $200/mo is about to be a line item in an S-1. curious how they'll frame the unit economics when heavy users cost multiples of what they pay. the IPO will force the free AOL CD phase to end. plan accordingly.

@alexocheema 8 solar panels, a starlink dish, and a big mac stack on the roof of a sprinter van. the republic doesn't need a data center.

Stacking Mac Studios is the new Mac Pro. Apple Silicon is the most advanced silicon you can buy + memory unit economics are better than anything else available today. High end market is now served by Mac Studios with RDMA over Thunderbolt turning all the Macs into one Big Mac.

@ClementDelangue @Pinterest @Airbnb @NotionHQ @cursor_ai @eoghan @intercom the API-to-self-hosted migration is the "cloud-to-on-prem" cycle all over again. starts with cost, ends with control. every company running agents at scale will eventually need to own their inference stack. "rented land"

After @Pinterest @Airbnb @NotionHQ @cursor_ai, today it’s @eoghan @intercom publicly sharing that they’re finding it better, cheaper, faster to use and train open models themselves rather than use APIs for many tasks. And hundreds of other companies are doing the same without sharing. Ultimately, I believe the majority of AI workflows will be in-house based on open-source (vs API). It took much more time than we anticipated but it’s happening now!

@shauseth the next one is building w/ agents right now

god i miss steve jobs. we are truly lost

Doom has been achieved internally.