Hacker Newsnew | past | comments | ask | show | jobs | submit | notphil's commentslogin

Cool idea. I can see this, if extended, being useful.

* A GitHub MCP exposing your code and issue contributions

* A site that exposes CV-data of candidates.

* An agent LLM iterating on all these, finding candidates that match roles.

Or vice versa, finding roles for a given candidate.

I might not be actively looking for roles, but I'd like to be aware of opportunities that might be a good fit. Recruiters historically have wasted my time.


I've met a few good recruiters, to be sure. But the median one definitely seemed to just match candidates up to roles in an entirely mechanical way, not even as well as an LLM could (because it at least would be informed roughly about whether or not experience in X translates at all to experience in Y, and not be tricked into thinking e.g. Java and JavaScript are at all functionally related). I wonder how those folks are doing these days, and how well they'll be doing in a few years.


> that the number of teams is such that phi is a good base

Any theories why this might be the case?

> 2. that the rounding all happened to go the right way.

+1. See how close some title races are: 2013-2014 comes to mind.


When there are more teams but the min/max skill is unchanged, they must be closer together, so the base is lower. When there are fewer teams, farther apart, the base is higher.

The number of teams is constrained by "we need enough teams so there's actually variety" and "we can't have so many teams that we can't keep track of them all".


2018-2019 and 2021-2022 were decided by a point, 2011-2012 on goal difference.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: