anotherjesse

prediction: resurgence in generative logo systems now that brand designers can easily vibecode them

@deepfates x.com/anotherjesse/s…

current setup: - $150 thinkpad touchpoint bt keyboard - $400 1080p glasses - DeX compatible phone/table device although I switch to a daylight tablet with same bt keyboard sometimes also testing using the even reality g2 glasses as a terminal that I control to send updates to glasses

If you assume you're going to do all your work from your phone in the near future, what is the best phone to have

@sawyerhood Making sloplings is interesting The way it allows experimentation is quite fun

my adventures in slop forking are making me question everything i once knew about software

Inspired seeing @graycrawford show some of his work from last decade to today. Exploring latent spaces non-verbally Deserves a full talk :)

@vikhyatk :nods: so say we all :sob:

remembering our hero

@ritwikpavan @EvenRealities @hwnation are you open for people using the device without the app? x.com/anotherjesse/s…

Hosting an @EvenRealities x @hwnation dinner in SF this Friday (3/20) Small group of hardware founders, engineers, and operators. A few spots left. Who wants in?

Woot! Web Bluetooth means we don't even need a local app - just connect to the glasses via a webpage @EvenRealities G2 glasses FTW

Where we're going we don't need apps! (progress on connecting/showing content on Even Reality G2 Glasses via BLE from macos / linux directly)

Woah some Swedes figured out actually-good glass printing. “Laser ironing” to flatten the layer lines, and a couple hundred um glass fiber as feedstock by the look of it. Great stuff, somehow around since 2024 and I just found them??

strange flowers of the mind are blossoming, silently and very fast.

The small new OpenAI models are great at runescape! For reference, GPT-5.4 Mini is 3x cheaper than Sonnet

@HamelHusain asking it to rank them / ability to file unrelated bugs (that go into a human review before uploading) -- anything less than medium / considered "next phase" gets skipped. I would say 90% of the things codex xhigh tags as medium+ are actually things I want addrssed

@HamelHusain I've had codex 5.4 xhigh for reviewer, high for coder do pretty good at this. also long as there is good "verification" loops that describe intended behavior

One thing that makes me feel that code factory has not arrived yet is the following experiment: 1.Ask a LLM to do an in-depth rigorous review of your code 2. In a new thread, as same/different LLM to consider those review comments independently and address issues it agrees with 3. Keep repeating until no new concerns I find that this loop always goes on for a ridiculously long time, which means that there is a problem with the notion of claude-take-the-wheel. This seems to happen no matter the harness or the specificity of the specs. It works fine for simple applications, but in the limit if the LLMs have this much cognitive dissonance you cannot trust it. Either this, or LLM are RLHFd to always find some kind of issue.

magnetic morph exploration

Love the exploration of how to think about generative ui/ideas

This but instead of time wolfram’s rule 30 pattern

@ryancarson amaze!

Taking my teenage boys to see Hail Mary in IMAX today. So freaking excited! :D

@dexhorthy does it count if they prep'd building context using tools based on analyzing previous conversations?

tool use as exception - any time the agent can’t give you a final answer, that’s an error case to catch and handle and retry