Tony Ge

vibe goader and gpu enjoyer

[Tony Ge @largePrawn]

so i lowballed 100s of luxury watch dealers in the USA using a voice AI and managed to negotiate $15k off MSRP for a Daytona

https://t.co/3eJ5E2lodV

[Tony Ge @largePrawn]

@stevibe This is massive!

Also an excellent use case for local llms since theres a mess of forms I'd like to automate but no shot I'm passing my ssn in

[Bleys Goodson @bleysg]

Since everyone is asking, I ran DeepSWE on MiniMax M3.

Here is the lowdown. 15 of 113 passed!

19 if you count the 1.5x overtime I gave just to see.

Full report: https://t.co/RglaGGablq

[Tony Ge @largePrawn]

@jun_song If you use docker at all - no. I need it for work and it's unusable. I end up having to tunnel into my desktop

[Tony Ge @largePrawn]

@loktar00 @carygolomb And...? share your work with the class

[Tony Ge @largePrawn]

@torychlamps @qtcinderella post physique

[Tony Ge @largePrawn]

@magonmetal @noxmetals i imagine some chinese manufacturing gigachad is looking at this like

[Tony Ge @largePrawn]

@zeroxjackson nah i just be typing "fix this shi rq"

[Tony Ge @largePrawn]

@ElevenLabs @TheRealStanLee "Surely the public will be thrilled to generate ai stan lee goon slop"

[Tony Ge @largePrawn]

@GavinSBaker I think says more about chinese inference eng

[Tony Ge @largePrawn]

@starmexxx room temp IQ take

[Tony Ge @largePrawn]

@muratkuralnedus @bentossell live demos prolly

[Tony Ge @largePrawn]

Imagine unintentionally getting hasan-pilled as what is basically a hired chatter

also what's stopping any of these guys from just copy-pasting the transcript and generating a report for fox

[Hasanabi productions (meme account) @HasanabiProd]

Fox News has now hired reporters to monitor Hasan Piker's Twitch stream all day.

[Tony Ge @largePrawn]

@sri9s forward deployed gooner

[Tony Ge @largePrawn]

@framer @work_louder Oh so it's come to this

[Tony Ge @largePrawn]

@jsawadd @latecheckoutplz @gregisenberg good first outcomes (since you asked in another comment)

wiring up a single MCP to your teams most used tool (slack or email maybe) to your teams other most used tool (claude probably)

[Tony Ge @largePrawn]

@jsawadd I literally did this for @latecheckoutplz/@gregisenberg

who are essentially all GTM folks

and by "this" I mean
- a second brain (black box of memory and content)
- 2nd Brain/Slack/Gdocs/internal-admin-panel MCPs for claude
- made fun onboarding videos for how to do that shi

[Tony Ge @largePrawn]

So very few tech folks I follow are also hasan followers.

just know i'll be looking at you sideways if any of us ever meet

[hasanabi @hasanthehun]

judge me by my enemies

[Tony Ge @largePrawn]

@mikeydsoftware @theo not to glaze theo or anything but bro what in gods name is this pfp

swe but cant resize an image?

[Tony Ge @largePrawn]

we've come full circle

a well written email is now signal

[vLLM @vllm_project]

Thanks to the community report, we recently identified a PR https://t.co/QWboSmskkF that attempted to solve a non-existent issue and was submitted as part of a “PR training” workflow for resume building.

The contributor involved has been banned from the vLLM community.

This kind of low-signal contribution increases maintainer review overhead and creates unnecessary operational costs for open-source projects.

As AI coding agents make generating large volumes of small PRs increasingly cheap, open-source communities will need to explore new ways to preserve contribution quality and reviewer trust.

While we are investigating how to deal with AI slop, we continue to highly value contributions from real users solving real production problems.

If you have an important contribution that has not yet received maintainer attention, please email us at:

[email protected]

Using a verifiable company or university email, include:
- your production or research use case
- the problem you encountered
- how your contribution addresses it

This helps us better prioritize impactful contributions while keeping the vLLM community open and collaborative.

As AI makes virtual contributors look increasingly real, authentic human collaboration matters more than ever.

vLLM’s mission remains unchanged: to make LLM inference easy, fast, and cheap for everyone — and we will continue working toward that goal.

[Tony Ge @largePrawn]

@no_stp_on_snek nah. 3 pcie slots is awful. he'll eventually need to replace his mobo + cpu