https://news.ycombinator.com/user?id=kirBurkhanovHacker News RSShttps://hnrss.org/hnrss v2.1.1Mon, 26 Oct 2026 12:55:20 +0000That calls lost-in-middle issue with LLM models. You have to use appropriate workflow patterns like map/reduce to handle big context. But be sure to run experiments on dataset first to understand what logic works well and stable. Where NOT! And how it can be combined to reduce LLM stohastic problems in prod

]]>Mon, 26 Oct 2026 09:18:38 +0000https://news.ycombinator.com/item?id=50241163kirBurkhanovhttps://news.ycombinator.com/item?id=50241163https://news.ycombinator.com/item?id=50241163Yeah, and the noise is also per model, run the same model five or ten times and put the confidence around it. Then try different models and different workflows. Thats actually how I do it when I build workflows and automations: test, test, and test, collect the statistic on where and what works better, then combine.

]]>Wed, 21 Oct 2026 09:48:04 +0000https://news.ycombinator.com/item?id=50178181kirBurkhanovhttps://news.ycombinator.com/item?id=50178181https://news.ycombinator.com/item?id=50178181