Top all-time · #24887
Thomas I. Liao
Personal blog · thomasliao.comaistartups
Thomas I. Liao's personal blog at thomasliao.com covers ai and startups. Since 2025, 3 different Hacker News readers have linked it 3 times, most often “Why eval startups fail (2025)”. It is still writing: the last post was on 2025-05-08, and you can follow it by feed.
Too few sampled comments (1) to say how often HN recommends it.?
Why HN links it
Linked posts are about why AI evaluation ("eval") startups fail. The only comment is the author's own definition of an AI evaluation as a prompt dataset, a scoring method, and a metric.
Summarised by Claude from a sample of 1 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Most-linked posts?
- Why eval startups fail (2025)2 people · 2 mentions · first linked 2025-09-06
What Hacker News says?
(Author) It's short for "evaluation", a test for an AI model. Specifically, an AI evaluation comprises (1) a dataset of prompts (as questions / tasks / queries), (2) some way to score model performance on each prompt, like a set of correct answers or a grading rubric that you can use with an LLM autograder, and (3) a metric, such as accuracy¹. (If you're already familiar with the term "benchmark", it's the same thin…
Comment linking /are_we_learning_yet.pdf
on “Why eval startups fail (2025)”Submitted Why eval startups fail (2025) as a story · 2 points
HN discussionSubmitted Why Eval Startups Fail as a story · 1 points
HN discussion