yohji1984

4 ms·

1.
Why My Open-Source Project Hasn't Done Better
4 points by yohji1984 2mo ago|1 comments
2.
by yohji1984 3mo ago
Sorry I will edit the post
3.
Claude Code team should try macro so users can complete 3x as many tasks
1 points by yohji1984 3mo ago|2 comments
4.
Is GPT-5.6 Sol Max Worth It?
2 points by yohji1984 3mo ago|4 comments
5.
Why people chasing after useless token saving plugins and ignoring real solution
2 points by yohji1984 3mo ago|0 comments
6.
Show HN: Agent use 83.1% fewer turns; 16.7 percentage points higher success(turaai.net)
2 points by yohji1984 3mo ago|0 comments
7.
That is why the tokensaving plugin and skill is completely useless(turaai.net)
2 points by yohji1984 3mo ago|0 comments
8.
Agent runtime reduces LLM turns by 80% with a higher success rate in DeepSWE(github.com)
2 points by yohji1984 3mo ago|1 comments
9.
by yohji1984 3mo ago
Hi HN, I've been working on an agentic runtime framework. The earlier benchmark eval is promising. But I can see the limits of the test design and the fragility of the runtime itself. I would like to ask for your reviews of the framewo
10.
by yohji1984 4mo ago
Sorry, I missed the Open Items section. You're right about that, designing a good eval harness can be difficult and expensive. Maybe we need some kind of community project for agentic evals, where people can share eval harnesses and ru
11.
by yohji1984 4mo ago
I'm wondering why all these token-saving solutions focus their benchmarks exclusively on simple Q&A tasks. If their tools truly saved money in real, long-term programming tasks, they would have definitely published those benchmark