Hacker News: aiagenta2z

New comment by aiagenta2z in "Qwen3.8-Max: A New Bar for Coding and Cowork"

aiagenta2z — Mon, 03 Aug 2026 15:30:12 +0000

Yeah, Qwen3.8-Max is the new Flagship model for coding and harness system and many other benchmarks are reaching equal performance as Claude and other close models. That's gonna drop the price of LLM in agent landscape a lot.

New comment by aiagenta2z in "Stateless MCP has recaptured my interest"

aiagenta2z — Mon, 03 Aug 2026 02:57:51 +0000

Yes That's true. The original MCP server recommend stdio, streaminghttp and other formats and earlier month in 2026, they recommend switch to streaminghttp way which is easier to support higher QPS and distinguish request using individual session ids and multiple connections (But the implementation always give wrong MCP request error). And now they provide a single HTTP request which make the access easier. Probability because the growing market of skills and clis direct access instead of MCP tax (long prompts) to complete for Agents access plugin market. So basically it might be the results of competition from other alternatives!

New comment by aiagenta2z in "LLMs and Xfwl4"

aiagenta2z — Mon, 03 Aug 2026 02:48:43 +0000

thank for the explanation!

New comment by aiagenta2z in "Show HN: Symbio self fine-tuning AI loop"

aiagenta2z — Mon, 03 Aug 2026 02:46:46 +0000

Yep, Thanks for your reply and it's helpful!

New comment by aiagenta2z in "Show HN: Symbio self fine-tuning AI loop"

aiagenta2z — Sun, 02 Aug 2026 03:44:19 +0000

So How do you prevent the continuously fine-tunes itself from drifting away from its original capabilities? The idea of training LoRA adapters from user corrections is interesting, and how do you balance personalization vs. catastrophic forgetting?

New comment by aiagenta2z in "Handbook.md shows that long policy documents do not reliably govern agents"

aiagenta2z — Fri, 31 Jul 2026 12:12:15 +0000

Hi I just read your methodology and I have a quick question about how the handbook are parsed and feed into the context window? The article mentioned that each handbook contains roughly 8K to 79K tokens of extracted text, and did the harness system use grep or search tools to find relevant chunks and feeds to the context, or did it just feeds all the pdf output to the model? There might by distribution bias between real world tasks e.g. Agents grep keywords from docs and only use the relevant chunks. So all the models of the overall pass@1 is relatively low compared to real world scenarios, that might not be the same precision that user experience when they actually handle the daily task? How did the benchmark bridge the gap?

New comment by aiagenta2z in "LinkedIn adds a 'Seems like AI slop' button"

aiagenta2z — Fri, 31 Jul 2026 11:57:31 +0000

Yes, The A Slop button is helpful if most of contents are generated by LLM and no real opinions or human knowledge. And that's probability not worth reading at all!

New comment by aiagenta2z in "Agent Skill to Force Docs in ASD-STE100 Simplified Technical English"

aiagenta2z — Fri, 31 Jul 2026 02:50:13 +0000

I have a question about how did the project evaluate the technical English performance compared to other skills/MCPs and with LLM w/o skills? In the Github repo, there is a summary of "measured: 6 Claude models × 8 tasks × 2 conditions, 96 runs", Does it means the percentage of violations in the words? If that's the case, the measurement might be a little bit strange weather the STE measure is already in the prompt? STE violations per 100 words ▼ 72.9% (every model won). The bench or evaluation should not be in the same prompt, like eval/test.

New comment by aiagenta2z in "Ask HN: What apps are you building?"

aiagenta2z — Thu, 30 Jul 2026 13:18:43 +0000

I forgot to leave the Craftsman Agent site link, and just append it here: https://craftsman-agent.aiagenta2z.com/app.

Now it supports various app to turn ideas(prompt/images) to design blueprint and finished products: 3D Generator, Lego Build, Minecraft Voxel Builder, 2D Perler Beads Pindou Builder. Would like to get your feedback!

New comment by aiagenta2z in "Cocoa prices are easing. So why is chocolate still so expensive?"

aiagenta2z — Thu, 30 Jul 2026 03:24:47 +0000

I used to intern at Mars chocolate factories supply chain, and are familar with the cost of products such as Ferrero. 1. Overall Inventory Cost: From the supply chain perspective, there might be months ahead of procurement before the actual production of chocolate, so the decrease in cocoa materials will not reflect the overall cost soon and there fluctuations. 2. Raw Cocoa materials costs only takes a few of overall chocolate product costs. The sales, distribution, retails are large portion of the overall cost. For a typical Milk chocolate, the estimated cocoa material cost share of final retail price ~5–15%, Even Dark chocolate with high Cocoas the materials cost will only takes ~15–30% of the overall product prices.

New comment by aiagenta2z in "LLMs and Xfwl4"

aiagenta2z — Thu, 30 Jul 2026 03:17:43 +0000

Hi quick question, xfwm4 has long development history. And what strategies do you use when modernizing an old C codebase without introducing regressions?

New comment by aiagenta2z in "Ask HN: What apps are you building?"

aiagenta2z — Tue, 28 Jul 2026 02:42:33 +0000

We are building an AI Agent Designer web app-Craftsman Agent which can turn prompts into ready to use 3D/2D design, such as a Tesla car wrap, a Perler Beads Pattern, etc. You can use "Generate a blue and white yacht", "Generate a Argentina Football Team Theme Car wrap or Perler Beads Pattern" to generate lego 3D blueprint, step by step assembly charts, etc.

New comment by aiagenta2z in "Kimi-K3 on HuggingFace"

aiagenta2z — Tue, 28 Jul 2026 02:38:56 +0000

Kimi K3 models are strong in coding and some abilities at a reasonable price. Nice job!