Job Hunting in 2025
In the words of Yogi Berra,
The future ain’t what it used to be.
I’m struck by how different job hunting is today versus seven years ago. In the past, I braced myself for the onslaught of emails and calls after initiating a job search. The biggest challenge was the logistics of taking an interview during the workday. I often took a walk away from the office to an area where I was unlikely to encounter coworkers. I apologized for all the background noise, explaining that I had to step away from the office to take the call.
For this job hunt, I put together my best ever resume based on the best yet version of me. I sent that canary into the coal mine, and it must have flown into a stalag… Wait, which ones go up and which ones go down? Regardless, I’m surprised by the lack of response.
According to Google, nearly 400,000 tech workers have been laid off since 2022. As a result, many job postings get hundreds of applicants. This sets up a challenge for both job seekers and employers alike. With such a strong tilt to an employer’s market, it’s overwhelming to choose between so many applicants. How can a hiring manager differentiate between hundreds of mostly similar resumes?
Enter automation 🎉! Wouldn’t it be amazing to have online systems that help us connect to potential employers? These systems could cut down on the redundancy with every job application, allowing you to focus on cover letters or questions about why you’d be a good fit! The systems would help hiring managers with efficiency, putting the best candidates at the top of the list. Everybody would win, and everyone would commute to work on the back of a unicorn. Right?
Not so much. No matter how many times I’m asked, my gender, my disability status, and my veteran’s status never changes. Yet I answer the same questions over and over, almost always with the same outcome. Either I don’t hear back, or I get a robo-mail about how much the team appreciates me taking the time to apply, though I’m not a good fit. In many cases, the poor fit remark leads to to conclude no human saw my resume.
Perhaps that’s the point, that the applicant tracking system (ATS) looks for the strongest keyword match between job applicant and job posting. Poor matches get discarded before the hiring manager has to read those resumes. A couple problems here. Say your resume lists relational databases as a skill, but the the job posting wants Postgres. While I admit that this is a contrived example, I’ve learned that my listed skills often don’t match with job skills due to wording choices. When I do hiring, I would consider this to be a skill match. Most problematic of all, bots are applying for jobs and competing with humans. I’ve read that bots sometimes act as an agent for human applicants, and the ethics of that are debatable.
Perhaps you’ve experienced this: you are asked to upload your resume to a job application. You hit submit, and sadly, your resume is minced into garbage. You appear to have a degree in Bash from Finite State with a GPA of 2019-2022. No matter how many times and ways I reformat the resume according to “ATS best practices”, it’s mincemeat! (I won’t name the worst resume parser, but it does rhyme with QuirkDay.) It is frustrating enough to make you quit applying for a while!
It’s not cool to talk about how great things were back in the day, like Drunk Uncle on SNL. But there are good lessons to take from the past - things worth bringing back. We used to try to get to know the person we considered hiring. An applicant was not a bullet list of skills that we would eventually probe for test anxiety in front of strangers. The best interview step ever was when a company asked me to give a presentation to the team about myself. It didn’t need to dive into tech; rather, it had to tell the team who I was. This was the make or break step in the process. No keywords, no live coding, no hypotheticals - just people getting to know each other.
Candidates are people asking you to take a risk on them. That has not changed. The process of connecting candidate to company has, and it is broken.
