Hacker News: sosodev

New comment by sosodev in "We moved Railway's frontend off Next.js. Builds went from 10+ mins to under 2"

sosodev — Wed, 08 Apr 2026 21:51:46 +0000

I don't know if Next.js, TanStack, etc are more abstract than Rails, Django, etc. They're undoubtedly more complex though. I also find it hard to believe that it's some sort of conspiracy by management to make developers more fungible. I've seen plenty of developers choose complexity with no outside pressure.

New comment by sosodev in "We moved Railway's frontend off Next.js. Builds went from 10+ mins to under 2"

sosodev — Wed, 08 Apr 2026 17:54:49 +0000

I think the unfortunate truth is the simplest. Web development has long been detached from rationality. People are drawn to complexity like moths to a flame.

New comment by sosodev in "Things I Think I Think... Preferring Local OSS LLMs"

sosodev — Thu, 02 Apr 2026 18:56:29 +0000

Qwen3-coder-next is way worse than Sonnet 4.5. Also, despite he lack of "coder" in the name Qwen3.5 is much better at coding than Qwen3-coder-next so you might want to check that out.

New comment by sosodev in "Qwen3.6-Plus: Towards real world agents"

sosodev — Thu, 02 Apr 2026 15:44:42 +0000

I don't know how well it performs, but you can extend Qwen3.5 to 1 million token context using YaRN. Also, Nemotron 3 Super was recently released and scales up to 1 million token context natively.

New comment by sosodev in "IronGlass Brings Legendary Soviet Cinema Lenses to Mirrorless Cameras"

sosodev — Mon, 30 Mar 2026 23:22:51 +0000

These prices are insane. You can buy all (most?) of the lenses they’re recreating for a fraction of the price and adapt them to a mirrorless camera no problem. I bought a Helios 44-2 recently for $100 and adapted it to my camera for like $15.

New comment by sosodev in "MSA: Memory Sparse Attention"

sosodev — Tue, 24 Mar 2026 22:10:54 +0000

I spent some time trying to understand this paper and I think calling this a new attention mechanism is a bit misleading. As a dead comment pointed out this is much closer to RAG. It's not exposing all 100M tokens directly to the model while doing each prediction. However, the RAG mechanisms have been integrated directly into the model architecture and that means it can have higher accuracy and lower latency. The higher accuracy is because it isn't storing text, but rather the actual in-memory representations (K/V, compressed tensor representations, routing keys, etc) of each document so it can search and utilize them more effectively. Given that it's computing up to 100x the context space it, like RAG, cannot process that volume in realtime. They explicitly state the the model needs to do offline encoding before handling inference. So you shouldn't expect to just send 100M tokens over an API and start getting a response.

I also think some of the benchmarks are misleading. Getting a RAG system to do an attention benchmark and then comparing it against a model without RAG just isn't fair. It is obviously better but it's not apples to apples. Some of the benchmarks compare against model+RAG and there the delta in performance is much smaller.

New comment by sosodev in "Apple Business"

sosodev — Tue, 24 Mar 2026 16:53:06 +0000

I think it's hard to know where to draw the line between derivative product and something unique. If we follow your logic that TSMC hasn't done anything new, then aren't all computer manufacturers just rehashing the ENIAC or whatever? Is a Tesla just a better model T? No, arguably we would say that these products are new to market because they've integrated new technologies in unique ways and often expended massive capital on R&D to do so. TSMC is no different.

New comment by sosodev in "Apple Business"

sosodev — Tue, 24 Mar 2026 16:26:48 +0000

TSMC. They dominate the semiconductor market because they're consistently first to market with the world's most advanced chip fabrication.

New comment by sosodev in "“Collaboration” is bullshit"

sosodev — Mon, 23 Mar 2026 18:50:06 +0000

Can we actually align incentives at scale? It seems to me that if it were possible we would live in a utopia.

New comment by sosodev in "Tinybox – A powerful computer for deep learning"

sosodev — Sun, 22 Mar 2026 15:28:56 +0000

Most people are using something in the llama family for inference. Llama server is my go to. Unsloth guides describe how to configure inference for your model of choice.

New comment by sosodev in "Tinybox – A powerful computer for deep learning"

sosodev — Sun, 22 Mar 2026 15:17:28 +0000

What models are you testing? A 120b model with hybrid attention should fit within 80gb of VRAM fine at a 4-bit quant. Also, 4-bit quants that are done well are generally fine. They certainly don’t make the model unusable.

Whole-Brain Connectomic Graph Model Enables Whole-Body Locomotion Control in Fly

sosodev — Thu, 12 Mar 2026 21:29:20 +0000

Article URL: https://arxiv.org/abs/2602.17997

Comments URL: https://news.ycombinator.com/item?id=47357394

Points: 2

# Comments: 0

New comment by sosodev in "New farm bill would condemn pigs to a lifetime in gestation crates"

sosodev — Mon, 09 Mar 2026 16:26:41 +0000

I didn't intend to. I think that domesticated animals have long had a harmonious relationship with humans so I find it a bit difficult to believe that it's always an ethical dilemma. Pets are just the most obvious lens to identify that.

I also think we need to be careful with the idea that we should entirely avoid suffering because it's impossible to do.

New comment by sosodev in "New farm bill would condemn pigs to a lifetime in gestation crates"

sosodev — Mon, 09 Mar 2026 16:18:48 +0000

I'm skeptical of this claim because there's clearly a growing population that hates the idea of putting anything they don't understand in their bodies. Genetically modified vegetables, food dyes, vaccines, etc.

I find it hard to believe you could convince a large portion of Americans to eat lab grown meat just to save a buck.

New comment by sosodev in "New farm bill would condemn pigs to a lifetime in gestation crates"

sosodev — Mon, 09 Mar 2026 16:13:55 +0000

Are all pets suffering?

New comment by sosodev in "New farm bill would condemn pigs to a lifetime in gestation crates"

sosodev — Mon, 09 Mar 2026 16:04:33 +0000

This is a false dichotomy. The choice is not lab grown or suffering. Farmed animals could live happy, healthy lives and then be culled in a humane way.

The problem is that it costs slightly more and our society is more concerned with cost than animal suffering.

New comment by sosodev in "How to run Qwen 3.5 locally"

sosodev — Sun, 08 Mar 2026 15:25:52 +0000

I’ve tried it via openrouter. It’s very good, but for some tasks frontier models are still significantly better.

For me, the 122b model is good enough on my own hardware that the downsides can be worked around for the sake of privacy and cost savings.

New comment by sosodev in "Something is afoot in the land of Qwen"

sosodev — Thu, 05 Mar 2026 01:36:28 +0000

I’ve been running it via llama-server with no issues. Running the latest Bartowski 6-bit quant

New comment by sosodev in "Something is afoot in the land of Qwen"

sosodev — Wed, 04 Mar 2026 20:44:12 +0000

Around 20ish tokens a second with 6-bit quant at very long context lengths on my AMD AI Max 395+

I’m trying to use local models whenever possible. Still need to lean on the frontier models sometimes.

New comment by sosodev in "Something is afoot in the land of Qwen"

sosodev — Wed, 04 Mar 2026 17:05:32 +0000

Some of the early quants had issues with tool calling and looping. So you might want to check that you're running the latest version / recommended settings.