Hacker News: seunosewa

New comment by seunosewa in "Elevated errors on Claude Opus 5"

seunosewa — Mon, 27 Jul 2026 12:45:33 +0000

GPT 5.6 Sol is so free of drama for me for brainstorming, writing prompts, etc. Tasks I always did with Opus until 4.7.

New comment by seunosewa in "Claude Opus 5"

seunosewa — Sun, 26 Jul 2026 11:41:52 +0000

Provide an option. An autonomous mode or an interactive mode.

New comment by seunosewa in "Claude Opus 5"

seunosewa — Sat, 25 Jul 2026 18:17:26 +0000

The problem is it should ask you before doing these things. This is a cherry picked example. There are times when they go on misadventures.

New comment by seunosewa in "Qwen 3.8"

seunosewa — Sun, 19 Jul 2026 15:49:11 +0000

They are trying to make money. That's what firms in any capitalistic economy care about the most. Regardless of the government's presumed interference, the companies themselves are all trying to make money. All competing for subscriptions and API payments.

One aspect of this is making a name for yourself i.e. PR. Making a capable model open source helps a lot with that.

New comment by seunosewa in "Grok 4.5"

seunosewa — Thu, 09 Jul 2026 06:41:39 +0000

How does Fable compare?

New comment by seunosewa in "Grok 4.5"

seunosewa — Thu, 09 Jul 2026 05:48:58 +0000

The input price is quite high. That's what gets you.

New comment by seunosewa in "Grok 4.5"

seunosewa — Wed, 08 Jul 2026 20:49:52 +0000

Not true. The only issue is cost of frontier models.

New comment by seunosewa in "The short leash AI coding method for beating Fable"

seunosewa — Sat, 04 Jul 2026 08:40:45 +0000

One of my favourite things to do is to ask the models, "what does this code do and why?" They are usually not far from the truth. From the perspective of a LLM that has been trained on all of the public code on GitHub, issues, and pull requests, your code and its intents are not that unique. They can see patterns in the code that point to the purpose. The constraints that shaped your code are the same constraints everyone else had.

Developer intents and aspirations aren't unique either. And in an LLM-written codebase, the chat history and prompts are available. The constraints, the rejected approaches, the goals. Users ask the models to write MD files that directly state their intents and aspirations for the ode.

New comment by seunosewa in "The short leash AI coding method for beating Fable"

seunosewa — Fri, 03 Jul 2026 14:14:54 +0000

We often don't understand the code we wrote 6 months ago.

New comment by seunosewa in "The short leash AI coding method for beating Fable"

seunosewa — Fri, 03 Jul 2026 10:41:04 +0000

You can just ask the model to explain the code to you.

New comment by seunosewa in "Qwen 3.6 27B is the sweet spot for local development"

seunosewa — Tue, 30 Jun 2026 03:22:11 +0000

You can get some work done by using low power mode even when plugged in, and making your fan start running when the temps just start to rise (maybe 40 degrees. Use a third party fan app to set it up

New comment by seunosewa in "VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO"

seunosewa — Tue, 23 Jun 2026 15:41:28 +0000

Then smaller the models are, the longer they have to reason when dealing with complex problems. The trade-off is real.

New comment by seunosewa in "We should be more tired than the model"

seunosewa — Fri, 29 May 2026 13:27:19 +0000

Unless you log its mistakes and how they were solved in decisions.log

New comment by seunosewa in "YouTube to automatically label AI-generated videos"

seunosewa — Thu, 28 May 2026 00:02:31 +0000

AI has completely ruined animal short videos on Youtube. Videos of pets behaving like humans are everywhere. At first they warm your heart, then you realize that you've been tricked.

New comment by seunosewa in "If AI writes your code, why use Python?"

seunosewa — Tue, 12 May 2026 14:37:23 +0000

What do you mean? Those are valid identifiers but programmers aren't required to use them.

New comment by seunosewa in "If AI writes your code, why use Python?"

seunosewa — Tue, 12 May 2026 14:22:20 +0000

Less verbose languages also use fewer tokens, saving precious context.

New comment by seunosewa in "Grok 4.3"

seunosewa — Fri, 01 May 2026 15:29:42 +0000

They dropped the output cost, butthe input cost is relatively high. This is a recent trend. Seen with DeepSeek 4 Pro as well.

New comment by seunosewa in "Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving"

seunosewa — Tue, 21 Apr 2026 04:30:13 +0000

API costs are really high compared to subs.

New comment by seunosewa in "Google releases Gemma 4 open models"

seunosewa — Sat, 04 Apr 2026 20:59:12 +0000

Now try to use it to develop a simple app.

New comment by seunosewa in "Axios compromised on NPM – Malicious versions drop remote access trojan"

seunosewa — Tue, 31 Mar 2026 15:35:39 +0000

We could have different Python package bundles: Python base. Python webdev. Python desktop.