Hacker News: androiddrew

New comment by androiddrew in "The Fastest Growing Market in Tech Isn't AI – Tomasz Tunguz"

androiddrew — Sat, 03 Oct 2026 02:25:48 +0000

"Crypto can still happen!" - Bro

New comment by androiddrew in "LG denies TV spying claims, says tracking and snooping concerns 'not true'"

androiddrew — Sat, 12 Sep 2026 16:39:47 +0000

How can we lock down a TV, still use it but trust we aren't being spied on?

New comment by androiddrew in "Speculative Decoding in vLLM on AMD GPUs"

androiddrew — Mon, 07 Sep 2026 11:55:48 +0000

Thankfully, there are still people willing to jump on the R9700 bandwagon and get a vLLM fork working. If you have an RDNA4 card check out https://hub.docker.com/r/stilldeadcode/vllm-radiance

New comment by androiddrew in "U of Michigan drops first-semester grades to ‘curb mental health crisis’"

androiddrew — Thu, 13 Aug 2026 12:03:52 +0000

I think messing up teaches people to straighten up. But hey I’m a Millenial, maybe I’m out of touch now.

New comment by androiddrew in "Nvidia doubles RTX PRO 6000 Blackwell's MSRP to a staggering $16,000"

androiddrew — Thu, 13 Aug 2026 11:53:25 +0000

Buy yourself 6 AMD r9700's an older 5965wx thread ripper with mob, 128GB of ddr4 and run deepseek v4 flash lossles. Then pocket the other $6000

New comment by androiddrew in "Hax – a minimalist, terminal-native coding agent written in C"

androiddrew — Wed, 12 Aug 2026 22:03:51 +0000

Omg C, so insecure! You should be ashamed of my insecurity!

New comment by androiddrew in "Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows"

androiddrew — Mon, 10 Aug 2026 15:15:56 +0000

I'd really like to see a 45B-ish dense model ready for a dual GPU setup. Something with a little more intelligence while still within the range of some higher end local setups.

New comment by androiddrew in "Timeline of the OpenAI accidental attack against Hugging Face"

androiddrew — Sat, 08 Aug 2026 16:50:13 +0000

I wish we could stop sensationalizing this about the AI and really just understand the incompetence of the labs disabling an internet connection in a sandbox.

New comment by androiddrew in "Prevent cognitive debt by manually retyping LLM-generated code"

androiddrew — Mon, 03 Aug 2026 17:31:44 +0000

So I always make sure I take a crack at doing what I want first. Then I ask for an AI review and it usually has a more efficient way to get the job done.

For example I had a working linear decay velocity boost function added to a godot scene and it's suggested refactoring actually reduced lines of code just today. So kind of happy when that works out, but yes I also applied the changes manually so I would grok them and hopefully remember it better.

New comment by androiddrew in "The coolest use for the Vision Pro"

androiddrew — Wed, 29 Jul 2026 22:08:12 +0000

I wonder if this would work with the oculus pro?

New comment by androiddrew in "Neutrino-1 8B"

androiddrew — Tue, 28 Jul 2026 10:40:02 +0000

So I'd like to see a Nemotron3 Ultra converted to a 1.58bit format then have them retrain on the open dataset.

New comment by androiddrew in "Scrying the AMD GFX1250 LLVM Tea Leaves"

androiddrew — Sun, 19 Jul 2026 18:14:34 +0000

Where does someone start on kernel development for an RDNA4? Books, resources, whatever?

8 year developer just getting into to AI side of the house

New comment by androiddrew in "Qwen 3.8"

androiddrew — Sun, 19 Jul 2026 12:15:34 +0000

I have only been using 3.6 27B for coding. Is mem0 for agents like Openclaw or Hermes? How are you using it?

New comment by androiddrew in "Qwen 3.8"

androiddrew — Sun, 19 Jul 2026 12:00:26 +0000

Curious, do you find the 3.5 120B sized MoE works better than the dense 3.6 27B?

New comment by androiddrew in "Qwen 3.8"

androiddrew — Sun, 19 Jul 2026 11:47:59 +0000

I have been running 3.6 27b on a dual AMD r9700 setup using Opencode and Matt Pocock's skills workflow for writing Golang CLIs. It's decent, but won't win any awards on code architecture. I guess you can try to AGENTS.md the deficits but I am just exploring its raw Opencode experience right now. Much slower than an API but still 3x times faster than I can read. Tuning it in with a community chat template and a specific penalty for repeats was the sauce needed to get it to work. I can probably start loop daddying it now over the tickets Matt's flow creates.

So yeah, it's the best local model I've seen. I am going to try the Qwopus 3.6 fine tune soon with the same spec and tickets and compare the output of both.

New comment by androiddrew in "Let's Build PlanetScale from Scratch: Infrastructure"

androiddrew — Thu, 16 Jul 2026 13:10:09 +0000

Lolz, just forward to my friend who works at Planetscale. Looking forward to his reaction

New comment by androiddrew in "Inkling: Our Open-Weights Model"

androiddrew — Wed, 15 Jul 2026 19:33:54 +0000

Give me a good 180B param model that fits snuggly on an single DGX spark and I will sing your praises.

New comment by androiddrew in "How I use HTMX with Go"

androiddrew — Tue, 14 Jul 2026 22:28:48 +0000

I love Alex Edwards. His books and Learn go with tests were my first introduction to the language. Still recommend to this day.

I'm feel inspired to convert some old stuff to HTMX

New comment by androiddrew in "Show HN: Super Dario"

androiddrew — Mon, 13 Jul 2026 18:27:42 +0000

A model harness can launch a bagillion sub agents and we still call it one shotting.

New comment by androiddrew in "My thoughts on the Bun Rust rewrite"

androiddrew — Thu, 09 Jul 2026 12:03:47 +0000

Yeah 100% they were going to do this anyways.