davosdo

Software Engineer

[davosdo @davosdodev]

@maria_rcks I feel like I lost five minutes of my life reading this

[davosdo @davosdodev]

Holy molly.

[Jacob Coxon @hilbertspaess]

I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.

[davosdo @davosdodev]

@ImLunaHey I think it depends on how much each person earns. In my case, we split it 80/20 because I earn more.

[davosdo @davosdodev]

Working on it! https://t.co/a6feQnh1dJ I think i'm gonna do a redesign to simplify things

[Benji Taylor @benjitaylor]

A personal website should feel like you’ve briefly left the rest of the internet

[davosdo @davosdodev]

@maria_rcks i hope!

[davosdo @davosdodev]

@gastonfonceadev Vas por Cursor o Codex?

[davosdo @davosdodev]

@maria_rcks @t3dotcodes for some reason when i have cursor cli and grok cli, the cursor one fails to load in t3code

[davosdo @davosdodev]

@suni_code Unreset

[davosdo @davosdodev]

Sad, but predictable because the grudge between Elon and Sam. The good news is Grok is getting so good that I doubt I'll even miss GPT Sol in Cursor

[OpenAI @OpenAI]

We’re ending our partnership with Cursor following its acquisition by SpaceX. Under our proposal, Cursor’s direct access to our models would end on November 12.

We know that the people most affected by this decision are the developers who rely on OpenAI models in Cursor. We care about their experience in this transition and we’re ready to go above and beyond to support them.

https://t.co/OzuCTzUjfX

[dax @thdxr]

lot of guesses on what ox alpha is but they are all wrong, kinda disappointed so just going to tell you

ox alpha is a new kind of llm that recursively updates a persistent latent state instead of reasoning entirely through tokens

this lets internal representations converge before anything is actually decoded

those attractors generate shards that encode transformations between latent states rather than the states themselves

at sufficient density these shards compose into metaparameters that dynamically alter the residual geometry of the model without changing its weights.

we built it because there was one thing simply too large to fit inside the context window of any existing model

your mom

[davosdo @davosdodev]

I almost burned through my weekly Grok Bot usage in two days. Most of it went into how I wake the agents, not the work itself.

If you're already on Grok Bot, how do you run it? https://t.co/zJ7pjn8HHx

[davosdo @davosdodev]

@poteto @bot I just learned this after i created lots of agents, channels and routines lol

[davosdo @davosdodev]

I’m genuinely impressed by the workflow combining @t3
Ox Alpha, and @opencode .

It’s not just about having a powerful model. The quality of the results, the overall ADE experience, and the tools around it make the entire development process feel next-level.

It's gonna be an amazing weekend using these tools 😃

[davosdo @davosdodev]

@maria_rcks nice bait

[davosdo @davosdodev]

After switching from Codex to Cursor completely, I haven't thought about going back once.

Cursor and Grok cover everything ChatGPT and Codex used to offer. SuperGrok makes the whole thing stronger. Throw in X and Grok @bot and this is the best AI coding setup right now.

I'm happy to spend $100 or more on it.

[davosdo @davosdodev]

Just cancelled Codex, i think i'm gonna go full with Cursor, the Ultra plan

[davosdo @davosdodev]

@theo Lakebed updates

[davosdo @davosdodev]

Lately, I’ve been reaching for T3 Code more often than Cursor or Codex.

Its UX feels intuitive and approachable. The inbox-style sidebar is the real standout and might be my favorite take on an agent sidebar so far. The theme switching is also incredibly well done.

One thing I’d love to see is a way to reuse skills configured in other harnesses, unless that’s already supported and I just haven’t figured out how to enable it 😅.

Beyond that, T3 Code already feels like a very capable ADE, especially for an alpha.

[davosdo @davosdodev]

@nanahrii Congrats on this milestone! I think i'm gonna use your app regularly

[davosdo @davosdodev]

@poteto @kenwuuuu it's enough for me