They're mostly interested in things I've already done in the past that involved large amount of users, which automatically filters out all those high agency projects.
Getting many users for your projects would be an option, but I see two problems there: - finding users requires a set of skills that is completely orthogonal to software development, so you'd need to learn those skills on the side - finding users is getting harder and harder, as your new project is the 100th release of the day
Besides, if you manage to build a good project and find users, you might not need a job at all. But let's say there's a 10% chance that happens. Then there's a 90% chance you go into your first interview as someone who failed to get a project off the ground, which is kind of the opposite of showing high agency.
And the odds of success are probably closer to 1% or smaller.
When scanning resumes, I would look for interesting projects and then on inteviews, drill down deep on one or more of them to see what the person actually did, how they think about it and how they worked with others.
Unless it's a senior position, I wouldn't care about the customer/business aspects or even if they were the creator/leader or not. I hired in one country where they had a rigorous math education but the professional projects weren't a good fit. But they had interesting things to say about the technical aspects of open source projects they worked on. That's better than having to project someone's aptitude. I also let my engineers assess technical ability, of course.
If they failed to launch at all, even with all the generous free-tier options out there, I'd agree that would be a concern. But at least in Silicon Valley and many tech circles, a failure is considered a badge of experience as long you gave it a proper try.
Yes, we now have to build orthogonal skills, but that is also kind of the point. Not only does it show adaptability and capability with AI, we need to be building proof of competency that could also become a source of revenue. Because I feel like these days counting on getting a job is like having hope for a strategy.
I totally feel the discouragement you're facing, and sorry if I make it worse, but this may increasingly be the only path left for us and at least you're farther along than most. Because a) what else could you be doing instead? and b) 300 paying customers is non-trivial and may even become a source you could grow.
The funny thing about people is even though anyone can, so few do.
That is what they are looking for.
Impossible to prove, trivial to lie about. Take note how many genuine startups lie about their recurring revenue or partnerships.
Let's say a candidate even sends you IRS statements, tax returns, corporate filings, bank transfer records.. How can you, as an interviewer, verify that those records haven't been fabricated? Or even if they are 100% legitimate, how can you verify that the clients who purchased their services haven't been coerced in some way? Aka getting a few dozen friends or family to subscribe to your app while never ever using it once. That candidate could show, say, $1000/month recurring revenue for 3-4 months, while having 0 genuine users. And just for the record, a LOT of startups use this method to boost their early traction. A lot.
From my experience of interviewing candidates, I can conclude, nobody cares about your pet projects.
I was one of the few, who'd even look at a candidate's CV more than 5 minutes before the call, let alone, follow links and explore candidate's GitHub.
The hardest part is getting past the first stage, because it has nothing to do with your actual experience or skills. Latter stages are a lottery too, but at least you know the rules and can prepare. Hence the advice somewhere above to get in front of people who are already in the industry – is the best advice.
Given that, I used to always make it a point to glance at the candidate's GitHub or pet projects and bring up any that seemed interesting, at least to make them feel seen.
But things are changing, and these are not pet projects. These are complete, working end-to-end products that (ideally) you've gotten actual users for. Maybe even paying users.
At which point, it could actually become a revenue source, and hence, an alternative path.