We desperately need a couple of additional coders to work with our tech leads. However, (perhaps surprisingly) having no additional coders is better than someon

We desperately need a couple of additional coders to work with our tech leads. However, (perhaps surprisingly) having no additional coders is better than someone who's a bad fit (sorta like it's better to be single than married to someone you cannot stand). It might be that you're hearing about job openings at companies which are large enough to have a marketing budget or happen to make a narrow range of products which you may use. Candidate and company both prefer candidates who are "walked in the door" by a trusted intermediary. However, for a company such as ours, we're small, our folks are heads-down on their work so don't spend much time at conferences, we sell to enterprises so you wouldn't hear of our product, and we don't use recruiters. There's probably new hiring forums/recruiters/technology stacks which would help this process, but I get a few such offers each week and how would I determine what works? We've gone through a number of approaches to interviewing and haven't yet found an interview "formula" which reproducibly gets to firm assessments. An interview (typically filled with talking) usually distinguishes between a "probably no" and a "probably yes". In contrast, a day or two of the typical back-and-forth while coding on real problems produces a highly accurate assessment...and enough insight to pay a premium for someone who could really help us get where we're going. However, interviews rarely include such real programming. As some other posts have mentioned, recruiting (and particularly the interviewing aspect of recruiting) is tough. While it's bad news for a company to hire the wrong person, it's also bad news for the candidate. Assuming you're not looking to hop from company to company, a good job is IMHO for 5+ years--a substantial portion of your productive years. Once you get a good sense a company is attractive to you (and vice versa), why not offer to spend a couple of days (e.g. a weekend) working with someone with whom you'd actually be working if you were hired...before you or the company commits? For our company, maybe we could pay you as a consultant so you could work on real code/real problems. No, don't ask for big dollar consulting fees--the benefit to the company is probably negative given your limited context and the oversight you'll require--this is just to reduce the likelihood of companies engaging in this practice without being actually interested in hiring you. It's a big investment from both sides, so this would be after achieving the "probably yes" outcome. For both the initial, superficial programming tests and optional 2-day programming projects, how about proposing to the hiring company how you'd like to be assessed? For example, our company develops single-page, collaborative editing and annotation applications for patent offices. If a candidate liked our company's people during the normal interview, tried our software, and proposed to add a feature to an open source project we use, the results would be a pretty clear indicator. You'd also be doing something you find interesting, since you suggested it. Our jobs are at https://angel.co/edyt https://angel.co/edyt