Onik Dev

Software Engineer, with a focus on innovating, not debating

[Onik Dev @OnikDeveloper]

AI Companies are kinda screwed.
Eventually there will be a LLM that’s small but powerful enough that no one will bother using AI companies API

Prediction: 1-2 years, people will run their own models on a single GPU. Won’t be surprised if gov gets involved

[Onik Dev @OnikDeveloper]

Mood. I’m the same when people watch me code lol

[Onik Dev @OnikDeveloper]

Brudda. The summer’s over now 😂

[Adam Chester 🏴‍☠️ @_xpn_]

Fuck you British summer 🖕

[Onik Dev @OnikDeveloper]

@LottoLabs @somewheresy How do you get this? Is it just llama.cpp that bad? I’m finding it hard to run vLLM simply because it breaks when trying to boot it up LOL

[Onik Dev @OnikDeveloper]

Fine fine. I’ll learn Zig lol
Good interview! Worth watching

https://t.co/AyxqDJoOoD

[Onik Dev @OnikDeveloper]

Ha! Had a good chuckle on this

[Kit Langton @kitlangton]

I pasted an api key into opencode and it told me to rotate it immediately.

Is this supposed to make it harder to read or something?

[Onik Dev @OnikDeveloper]

Nice. Gonna buy this. The fact that I can run Linux. So many use cases. Nicely done flipper staff!

[Flipper Devices @flipper_net]

We're finally ready to talk about Flipper One — a project we've been grinding on for years and have rebuilt from scratch several times. Read blog post >>
https://t.co/tTywwytXzO

[Onik Dev @OnikDeveloper]

@Mayhem4Markets Yeah it seems like CUDA is going great. I’m rocking RTX 6000 Ada Generation and it’s doing well for me. Can’t wait to move to the pro series soon

[Onik Dev @OnikDeveloper]

@CtrlAltDwayne

[Onik Dev @OnikDeveloper]

Good. Competition is key

[CryptoGoos @cryptogoos]

🚨 THE MEMORY CARTEL IS ABOUT TO FALL.

Ex-Samsung chip boss says heavy Chinese investment in the memory market could crush the 414% DDR5 price spike within a year.

Goldman calls it RAMageddon.

Samsung, SK Hynix, and Micron control 70% of global DRAM and pushed prices from $6.84 to $27.20 in 3 months.

Now China is gearing up to flood the market.

Cheap memory = cheap AI compute = the cartel cracks.

[Onik Dev @OnikDeveloper]

@ChShersh

[Onik Dev @OnikDeveloper]

Crap. If Senior’s are leaving. How on earth are we gonna learn?

[Mario Verbelen @MarioVerbelen]

I'm leaving coding behind me, Bjarne is fully right about AI and if you want performance on the backend you should stick with the manual art of coding.

Management needs AI to be in every slide, so in the coming few years there the focus will be all into AI.

As a senior my brain refuses to validate AI output, it's way worse than guiding a junior in the team.

So my time in tech will end this year, and I'm not sure what my next step will be.

[Onik Dev @OnikDeveloper]

This new generation that’s coming to the workforce are gonna change the world. I can feel it

[ℏεsam @Hesamation]

it’s really sad how a revolutionary technology with the potential to cure cancer, push forward science, and transform humanity, is now HATED by the public because of
> corporate greed
> companies AI-washing layoffs
> and Dario’s fear mongering to sell AI

[Onik Dev @OnikDeveloper]

@LottoLabs @keennay I guess your 3090s are safe and sound then 😂

[Onik Dev @OnikDeveloper]

@martinromanuk @scaling01 It’s flash. So I’m assuming it’s due to the speed

[Onik Dev @OnikDeveloper]

Skill issue. Clean up your test properly

[Error 404: Humor Found😉 @expensiveaccide]

joys Of Automated Testing

[Onik Dev @OnikDeveloper]

@Alibaba_Qwen @arena Let’s see what they release. Cannot wait!

[Onik Dev @OnikDeveloper]

@MgClutchh Stop hacking 😂

[Onik Dev @OnikDeveloper]

Good. Make this the norm. Local AI should be the norm. Do not let big corporations take it away from us

[David Hendrickson @TeksEdge]

💥 What is this beast? Skymizer HTX301 is LLM 🛸👽👇

One PCIe card w/ 📦 384 GB memory ⚡ 240W TDP 🧠 Runs 700B LLMs locally

Vs NVIDIA RTX 6000 Ada:
48 GB • 300W • ~$7,500

Vs RTX PRO 6000 Blackwell:
96 GB • 600W • ~$8,500

HTX301 delivers 8× memory at less than half the power and specialized LPU inference beast for on-prem AI.

🔥 No clusters. No NVLink. Just plug & infer.
Pricing TBA • Early access open now

[Onik Dev @OnikDeveloper]

As initially expected.
The large AI companies are screwed.

The better the open source models are, the more they lose their value.

It’s no longer an intelligent problem. It’s a compute problem.

And because we’ve always had GPUs. Every household should have one to run local

[0xSero @0xSero]

I want my 6000s to lose all their value, that means my vision has come to life.

Everyone should be able to have fun.

[Onik Dev @OnikDeveloper]

Move over the Linux and degoogle. It’s worth it

[Brave @brave]

Exciting week for people still using Big Tech browsers