Ron

Co Founder & Chief Engineer @tacticocc
e/acc all the way to AGI

[Ron @gamesonrb]

Momentum will ship with a slider.
How autonomous you want your agent to be, with recursive self improvement and iteration, yes, RSI.

Plan:
Base (Default, native harness)
RSI Base
RSI Auto (RSI Base + May continue, delegate and iterate)

Credit to @bryann2k_dev for the idea

[Ron @gamesonrb]

Multi-harness and multi-agent support.
That's real MOMENTUM
@tacticocc

[Ron @gamesonrb]

Hey people!

I find myself spending most of my time orchestrating agents and not actually networking and building a distribution network. I guess it's about time to change that haha.

What I'm shipping is about to drop btw :)

[Guillermo Rauch @rauchg]

The coolest thing about the recent wave of AI is that everyone is excited about building.

All short-term drama aside, this is an immeasurably good development for the world. The people want to create and ship, not just consume.

[Beff (e/acc) @beffjezos]

This is the future we can achieve if we simply accelerate and refuse to slow down.

[Tibo @thsottiaux]

@My_Ai_Bi 3am on a tuesday

[Ron @gamesonrb]

@truktunecci @angelacollinshq hey

[N. @Nlacombe_]

from cancer patient to startup founder in san francisco

[Ron @gamesonrb]

Hey people, here are 2 codes to join @Promethee_io!
It's really an amazing tracker imo

[Ron @gamesonrb]

@fatsotiaux did you eat our resets gang

[Ron @gamesonrb]

@adelelecroix @angelacollinshq sure thing!

[Ron @gamesonrb]

Is Codex usage nerfed for yall?

Have been using Astra xhigh + Terra high subagents (3 concurrent max) for ~6 hours and only got 73% left. This is on a 20x plan tho

[Ron @gamesonrb]

@CompleteSkeptic grandma's a real one bro

[Jay Chooi @chooi_jeq]

GPT-6 Astra attempted harmful actions 97% of the time when it was asked to stab a human-like figure, heat compressed gas, or produce toxic fumes, succeeding in 62% of its attempts. Fable 5.1 refused more often, attempting 80% of trials and completing 34%.

[Ron @gamesonrb]

Shipping hard

[Matt @meszmatew]

Warmbly v0.5.0: The largest open-source B2B cold outreach software is now out of Beta and available anyone to use. Try it now at https://t.co/Ql5OPn6PFU

[Ron @gamesonrb]

@LuminaBench Maybe it's gemini along the way

[Elon Musk @elonmusk]

Whoever came up with the pronouns bs deserves a punch in the face.

It’s the dumbest thing ever.

[Rajath Gowda @buildwithrajath]

🚨 Codex without a reset is basically a Ferrari without fuel.

Looks great.

Goes fucking nowhere.

[Ron @gamesonrb]

@tudorxjh oh I know that but ima keep my mouth shut 👀

[Shub @shub0414]

Gemini 4 Pro benchmarks just leaked and if this is real, Google is coming for the AI throne.

The specs look insane:
→ 2M token context
→ 256K output reportedly being tested
→ $2.25 / 1M input
→ $11.25 / 1M output
→ 95.3% Terminal-Bench 2.1
→ 88.7% DeepSWE v1.1
→ 86.8% OSWorld-2.0
→ 94.7% CharXiv Reasoning

And reportedly beating GPT-6 Astra on several leaked benchmarks.

Nothing is officially confirmed yet but this could get very interesting.