I was waiting for this rambling post to get to the point until I realized it's just an ad for the author's new ATS. They're trying to convince you that other products are bad but theirs is good.
author here! i originally intended this to be a technical write-up on everything i learned building an ATS (as someone who's in HR). the original title was "how to build an ATS and why you probably shouldn't". i debated mentioning what i built, but it felt hollow to _ramble_ about schemas, architecture, components with nothing to show for it.
the more i wrote and reflected, the more i thought about why the market never corrects itself despite the tools being expensive and badly designed. i've worked with hundreds of recruiters and most use spreadsheets. that's not a workflow quirk but i think an indictment of something bigger which traces back to everything in the post -- the buyer who never uses the product, the integrations racket, the "AI-native" tools bolted on top of a broken foundation, etc. etc.
so i ended up writing the first half. it's drawn from my experiences frustratingly buying an ATS for a small business, and watching the dysfunction of procurement/integration/lack of adoption play out at the enterprise level.
admittedly, HR/recruiting tech is a very niche audience, so the technical section probably lands better with engineers who've been handed a recruiting project than with anyone actually working in HR. so i wanted to offer a resource from that perspective.
I think there's just an plain faulty assumption that hiring is optimized for best candidate in lowest time for the optimal work done delta improvement.
Hiring is a process with many different motives. Like:
- Signaling company growth
- Appeasing overworked employees that something is being done
- Signaling that you or your team is important by gatekeepimg the role
- Signaling that you are important by participating or contributing to the hiring process
- Endlessly window shopping candidates simply because finding the perfect one is fun
There's a simple fact that if no one is pressuring hiring to pick someone sooner, there is simply no motivation to. And hiring is everyone involved. Managers, engineers, c suite, anyone with a veto in the process of a candidate. A single kink in the pipe can drag on the process forever. Even if engineering is slammed, if the recruiter screen or even the final CEO interview doesn't interalize that, the process is borked.
Now the real question is where are the hiring platforms that optimize for these weird motivation. I bet a platform where you swipe candidates for fun and encourage the whole team bikeshed screener quizzes would do gangbusters. Straight up make it a company tinder where unless recruiting, engineer, and CEO all swipe right on a candidate its a match! (Barf)
I hate to be that guy, but HR is one of the things I always point to as a perfect example of "A system's purpose is what it does"
- HR's task is NOT with maximizing results/IC output
- HR's task is minimizing corporate risk
HR is, in most corporate environments, doing exactly what it is intended to do (minimize risk)!
Hiring anybody, from an org's perspective, is insanely risky for a million different reasons. Therefore, there are a million different (valid and invalid) reasons to reject a candidate - which is what overwhelmingly happens, unless HR is sidestepped via referrals and networking.
But does it minimize corporate risk? Those who get ghosted or face an unfair interview can overwhelmingly report a negative experience online, which then slowly drags the company down because it hurts the candidate pool. I assert it does not minimize this dimension of corporate risk.
Some interesting insights, but author is speculating on a technical solution for a process that's broken * for the job seeker *.
The fact that there's often thousands of applicants for one job is exactly what companies and recruiters want. This system shifts all the power to them, and they're perfectly happy with it. No amount of technical fixes will change this, or if it's even necessary.
the candidate experience angle is definitely worth its own post because i don't think it's black and white. plenty of companies get thousands of applications and are still dissatisfied with the talent quality (although i'm skeptical of this, i don't dismiss it entirely). others get thousands and simply can't process the volume, with 75% left unreviewed (ask me how i know).
if you work in big tech and you've ever referred an extremely qualified friend who's also in big tech and watched it get ignored despite 10 open support engineer or 20 account executive roles they'd be perfect for, that's exactly the incentives and infrastructure problem i'm describing.
AI might help with these problems for both recruiters and candidates though. pre-AI the pitch for many recruiting tools was about skills and keywords (garbage in, garbage out). but now we can use LLMs to reason positive signals related to qualification (compared against the job description, does resume/application show experience match based on years, sales quotas, industries, programming languages, etc.). the goal is surfacing more qualified candidates for human review, which matters a lot when a recruiter is literally waiting 5-10 seconds for one application to load in workday while every incentive they have pushes them to just source someone off linkedin instead.
My side project is in recruiting. We have both internal and external recruiters as our advisors. I've interviewed many more.
As one put it, "I can leave a developer job open for a couple of hours, and I'll get a dozen that can do the job. If I leave it open for a day, I'll get a hundred."
I was waiting for this rambling post to get to the point until I realized it's just an ad for the author's new ATS. They're trying to convince you that other products are bad but theirs is good.
Which seems weird, it's very technical. Monolith design, relationship types, I've never met an HR person who wondered about those kinds of things
author here! i originally intended this to be a technical write-up on everything i learned building an ATS (as someone who's in HR). the original title was "how to build an ATS and why you probably shouldn't". i debated mentioning what i built, but it felt hollow to _ramble_ about schemas, architecture, components with nothing to show for it.
the more i wrote and reflected, the more i thought about why the market never corrects itself despite the tools being expensive and badly designed. i've worked with hundreds of recruiters and most use spreadsheets. that's not a workflow quirk but i think an indictment of something bigger which traces back to everything in the post -- the buyer who never uses the product, the integrations racket, the "AI-native" tools bolted on top of a broken foundation, etc. etc.
so i ended up writing the first half. it's drawn from my experiences frustratingly buying an ATS for a small business, and watching the dysfunction of procurement/integration/lack of adoption play out at the enterprise level.
admittedly, HR/recruiting tech is a very niche audience, so the technical section probably lands better with engineers who've been handed a recruiting project than with anyone actually working in HR. so i wanted to offer a resource from that perspective.
I think there's just an plain faulty assumption that hiring is optimized for best candidate in lowest time for the optimal work done delta improvement.
Hiring is a process with many different motives. Like:
- Signaling company growth
- Appeasing overworked employees that something is being done
- Signaling that you or your team is important by gatekeepimg the role
- Signaling that you are important by participating or contributing to the hiring process
- Endlessly window shopping candidates simply because finding the perfect one is fun
There's a simple fact that if no one is pressuring hiring to pick someone sooner, there is simply no motivation to. And hiring is everyone involved. Managers, engineers, c suite, anyone with a veto in the process of a candidate. A single kink in the pipe can drag on the process forever. Even if engineering is slammed, if the recruiter screen or even the final CEO interview doesn't interalize that, the process is borked.
Now the real question is where are the hiring platforms that optimize for these weird motivation. I bet a platform where you swipe candidates for fun and encourage the whole team bikeshed screener quizzes would do gangbusters. Straight up make it a company tinder where unless recruiting, engineer, and CEO all swipe right on a candidate its a match! (Barf)
> The product is the sales demo that impresses VPs. Meanwhile, recruiters are still shuffling candidates around in Google Sheets.
This gave me a chuckle, because a colleague who talked with HRs just told me exactly this last week.
I hate to be that guy, but HR is one of the things I always point to as a perfect example of "A system's purpose is what it does"
- HR's task is NOT with maximizing results/IC output
- HR's task is minimizing corporate risk
HR is, in most corporate environments, doing exactly what it is intended to do (minimize risk)!
Hiring anybody, from an org's perspective, is insanely risky for a million different reasons. Therefore, there are a million different (valid and invalid) reasons to reject a candidate - which is what overwhelmingly happens, unless HR is sidestepped via referrals and networking.
> "A system's purpose is what it does"
POSIWID: https://en.wikipedia.org/wiki/The_purpose_of_a_system_is_wha...
But does it minimize corporate risk? Those who get ghosted or face an unfair interview can overwhelmingly report a negative experience online, which then slowly drags the company down because it hurts the candidate pool. I assert it does not minimize this dimension of corporate risk.
Some interesting insights, but author is speculating on a technical solution for a process that's broken * for the job seeker *.
The fact that there's often thousands of applicants for one job is exactly what companies and recruiters want. This system shifts all the power to them, and they're perfectly happy with it. No amount of technical fixes will change this, or if it's even necessary.
the candidate experience angle is definitely worth its own post because i don't think it's black and white. plenty of companies get thousands of applications and are still dissatisfied with the talent quality (although i'm skeptical of this, i don't dismiss it entirely). others get thousands and simply can't process the volume, with 75% left unreviewed (ask me how i know).
if you work in big tech and you've ever referred an extremely qualified friend who's also in big tech and watched it get ignored despite 10 open support engineer or 20 account executive roles they'd be perfect for, that's exactly the incentives and infrastructure problem i'm describing.
AI might help with these problems for both recruiters and candidates though. pre-AI the pitch for many recruiting tools was about skills and keywords (garbage in, garbage out). but now we can use LLMs to reason positive signals related to qualification (compared against the job description, does resume/application show experience match based on years, sales quotas, industries, programming languages, etc.). the goal is surfacing more qualified candidates for human review, which matters a lot when a recruiter is literally waiting 5-10 seconds for one application to load in workday while every incentive they have pushes them to just source someone off linkedin instead.
> The fact that there's often thousands of applicants for one job is exactly what companies and recruiters want
Are you sure that's true? I often read complaints here from hiring managers that have to wade through far too many obviously unqualified applications.
Yes.
My side project is in recruiting. We have both internal and external recruiters as our advisors. I've interviewed many more.
As one put it, "I can leave a developer job open for a couple of hours, and I'll get a dozen that can do the job. If I leave it open for a day, I'll get a hundred."