Hacker News: lhk931122

New comment by lhk931122 in "Research acceleration: The view inside OpenAI"

lhk931122 — Mon, 07 Sep 2026 02:42:52 +0000

Ah, success rate here are scored by an agentic classifier. And uncertain outcomes are excluded from the graph. The thing measured and grading it comes from the same house. In my setup, review agent pass work that an outside critic later rejects

New comment by lhk931122 in "Which tools do Claude, Codex and Cursor choose? We measured 17k runs to find out"

lhk931122 — Fri, 04 Sep 2026 04:20:21 +0000

I only get web search from Claude Code when I ask, only one of my last 8 sessions accessed the web at all. But they were the case of Opus here. Curious what Fable 5.1 model does instead of Opus.

New comment by lhk931122 in "Show HN: A lightweight, stateless database for agent memory"

lhk931122 — Mon, 31 Aug 2026 00:24:05 +0000

I tried to make a memory system in Claude Code (make it behave like a human from the perspective of memory) using markdown files and indexing plus hooks. But, retrieval is not as good as a human and since keywords are ambiguous, it doesn't behave as intended.

New comment by lhk931122 in "Ask HN: Why are Claude models so verbose?"

lhk931122 — Fri, 28 Aug 2026 13:01:50 +0000

We can control it (claude.md and plus use 'rules') but, it would be better if not verbose before I set it. yeah