San Francisco, CA
- Tweets 2,146
- Following 113
- Followers 136
- Likes 3,261
If you love what you do, you'll never work a day in your life
100-hour workweek challenge
Day 5: 15h
At the end of the day, my Uber driver asked me, “What’s your hobby?”
I said, “I like work.”
He laughed happily, like I was joking with him.
so I said, “believe me, I really like work, haha.”
The way I feel about work is actually very similar to how I feel when making art. I could spend the whole day, sometimes days, completely absorbed in it, entering that state of flow.
You may feel physically tired, but mentally fulfilled.
4
Hey Muse, turn the AC on and let's get to work!
sorry @alexandr_wang nobody is building muse gadgets today
it's 80 degrees in sf
1
15
it's easy to fall into the trap of micromanaging your agents instead of correcting the environment that shapes its behavior.
pstack 0.15.9 now ships a new /correct skill: if you keep correcting agents for the same mistakes, it finds the pattern and fixes it with architecture, types, and checks.
i like to imagine that commits rapidly flowing into the codebase are like a bonsai tree rapidly growing in whatever direction it pleases. what you want to do is not to give in to the chaos, but tame it intentionally with the right constraints
0.15.9 also has improvements to:
• /architect: now includes instructions on designing agent friendly architecture
• a new /benchmark-checklist skill based on @brendangregg's benchmarking checklist
here's a prompt you can give to grok bot to start architecting a better codebase with pstack:
/poteto-mode create a new Project agent to refactor and rearchitect my repo so that our architecture is more agent friendly. it should use both the /correct and /architect skill to look through past commits and find the most common pitfalls that agents fall into, whether its with code quality, performance, or bugs. if relevant you can also use the /recall skill to find past conversations for context. it should make a high level plan for me to review. it should also answer its own open questions by prototyping rather than defer to me. come back to me with a plan backed by real data. once i approve the plan, use either the stack or full autopilot playbook (ask me which i want) to execute
every time you intervene and correct your agent, you should think about how to eliminate it entirely.
in order of value:
1. categorically eliminate the problem through better architecture or choice of data structures
2. turn it into a lint rule or test so CI catches it
3. turn it into a skill or rule
4. have humans review the code to catch it (ngmi)
136
146
2,988
109,754
This is cool
🚨 MUSE ECOSYSTEM ALERT 🚨
today we are announcing Muse Gadgets!
this is an open-source ESP32 firmware and Linux SDK for anyone to make hardware that works with Muse
we are also releasing our own gadget—Muse Home Link—to enable your muse to work with your smart devices (TV, speakers, etc.)
12
here's how i shipped 2,500 PRs last month to production
this was originally supposed to be for Cursor Compile in London. i couldn't make it since i was livestreaming for Grok @Bot Galaxy so i'm making it available for free here on X! watch it on 2x speed, i talk slowly
528
1,234
14,701
3,003,513
this is going to be very interesting
Next week, @mattyp @roshan_s and myself will be speedrunning building a new company using only Grok @Bot on the Grok Bot Galaxy livestream! We’ve been given just 3 days to do it, or else we might be turned into pizza 🤪 it will be so fun and I hope to see you on the livestream!
I will have to miss the Cursor Compile conf in London and Amsterdam though. I will instead record an extended version of my talk and make it freely available here. So stay tuned for that post!
x.ai/galaxy
65
this is cool
it's here! i've been using Projects for a while and have shipped thousands of PRs with them. pstack works incredibly well with it, with your coordinator managing and orchestrating hundreds and even thousands of agents for you
these all work in the cloud too, so you can parallelize at an incredible scale. with Projects you can create your own personal software factory (or Michelin kitchen as i like to call it) without all the plumbing. just prompt your agent!
1
15
forming a clear thought and putting it into words is perhaps the biggest challenge for most humans
everyone spent ten years learning to code and it turns out the terminal skill was english composition. the machine will do anything you tell it. the only thing left between you and what you want is whether you can form a clear thought and put it into words
47
adding all of them to my collection lol
meet my new grok @bot tinkabot (v0.1.0)! she helps you make high quality grok bot plugins. you can ask her to look at your APIs, create an MCP and/or skills for it, and then wrap that up as a plugin that you can then submit to us for approval.
once it's approved, your plugin will be available for all grok @bot users to use!
let me know if you run into any issues, i've used it on some small toy APIs but would love to see how well it works on real ones
x.ai/bot/br5f3C4mc75QCMEHasz…
71
Dr Eggbot part of my crew now
try my Dr Eggbot (v0.1.0): x.ai/bot/93gOz3op1UQdBdbekQF…
he makes high quality bots for you! coding bots use pstack, non-coding ones have similar rigor
1
132
just checked the skill and this opens up a lot of possibilities
i also made a new skill called "/Make Bot UI" in pstack grokbot://app/v1/plugin/add?id=9717366
this teaches your bots how to make a custom UI that can interact with bots! for example you could use this to make a small app that gives you a bespoke UI that helps you go through all your emails in your inbox in a IG style interface, and each swipe sends a message back to your bot to do something "agentic" - like draft a DM to a creator, or to share it on slack.
38