WeWorkBuddy

Ask HN: Is post-training make AI models dumber?

Post-training is making models better at benchmarks. They seem more consistent at coding, but dumber to me and I want to know what y'all are noticing. Here are the big three I'm noticing (mostly in claude): 1) Agents cannot write natural language well. Claude uses vague phrases to avoid any technical term such as "What the caller calls" instead of saying "callee". 2) Agents cannot manage encapsulation; they always make spaghetti code. 3) Agents are afraid to break anything. T

Read original source
Source reference for Ask HN: Is post-training make AI models dumber?. Screenshot not yet available.
Source record · preview pending · news.ycombinator.comOpen the source for the full page.

Source record

Recorded metrics

points · 3 votes
2026-11-16 04:58:29.581Z
cumulative since publication

Research analysis

Reported problem

Source account: Post-training is making models better at benchmarks. They seem more consistent at coding, but dumber to me and I want to know what y'all are noticing. Here are the big three I'm noticing (mostly in claude): 1) Agents cannot write natural language well. Claude uses vague phrases to avoid any technical term such as "What the caller calls" instead of saying "callee". 2) Agents cannot manage encapsulation; they always make spaghetti code. 3) Agents are afraid to break anything. T

Frequency & alternatives

Recorded author: CleverLikeAnOx. Source votes: 3 votes (Nov 16, 2026 UTC; cumulative since publication); previous same-scope observation: 3 (Nov 15, 2026 UTC). These observations do not establish demand growth. Recent records to compare: “Ask HN: Tips for maintainability of large software projects using agents?” (Hacker News, Nov 15, 2026); “Ask HN: What AI agent infrastructure are you using in 2026?” (Hacker News, Nov 15, 2026). Shared themes are navigation hints; they have not been verified as the same need or independent corroboration.

Demand hypothesis

Treat the reported task or obstacle as an interview lead. A feature request, complaint or search for alternatives does not by itself identify a budget owner or willingness to pay.

Commercial evidence gaps

Missing evidence: distinct affected users, repeated occasions, current workaround, measured time or money cost, purchasing authority and a response to a concrete offer.

Validation decision

Group original reports by the same task and deduplicate authors and reposts. Ask how the task is handled today and what it costs; test a defined offer before claiming paid demand.