Building a $4M Empire When Everyone Said AI Would Replace Recruiters

The AI doomsayers had it all figured out. Recruiting would be automated. Resumes would be parsed by algorithms. Human recruiters? Obsolete.

Then Priyanshu went and built two companies doing $4M+ in annual recurring revenue. In the exact space AI was supposed to kill.

His secret? Understanding that AI doesn't replace human judgment—it amplifies it. And knowing the difference between what sounds smart in a tech article and what actually works when you're trying to match the right person to the right job.

This is the story of building in the eye of the AI storm.

Key Takeaways

• AI augments, doesn't replace: AI handles the tedious screening work, but human judgment is still critical for evaluating cultural fit, soft skills, and career potential
• Calculated risks over blind faith: Priyanshu bootstrapped for 2.5 years before taking the leap full-time, ensuring product-market fit before burning bridges
• Resume optimization is broken: Most "ATS-friendly" advice is outdated—what matters is relevance to the job description, not keyword stuffing
• Skill development beats career hopping: Building deep expertise in one area creates more value than chasing every new opportunity
• Sales fixes everything: A great product that nobody knows about dies. Learning to sell is non-negotiable for founders
• The recruiting industry is ripe for disruption: Most recruitment happens through networks and referrals, not traditional job portals—there's massive opportunity in making this process more efficient

The Full Conversation

You were working a comfortable job in the UK. What made you think "now's the time to jump into entrepreneurship full-time"?

I didn't just jump. That's the thing everyone gets wrong about startup stories.

I started Sprinx and Agylex while I was still working. Kept my job for 2.5 years while building on the side. Made sure we had revenue, made sure we had clients, made sure the thing actually worked before I took the leap.

The tipping point was when managing both became physically impossible. We were growing fast enough that I couldn't give either thing my full attention. That's when I made the call.

But it wasn't blind faith. It was a calculated risk backed by 2.5 years of proof that this could work.

Hot take: Most startup advice tells you to "just quit and go all in." That's terrible advice unless you've got proof of concept or a massive runway.

Tell me about Sprinx and Agylex. What's the difference between the two?

Sprinx is the recruitment side. We work with about 30 clients helping them fill technical roles. Think of us as the talent acquisition arm for companies that need engineers, product managers, technical roles.

Agylex is more on the career development side. We help individuals with resume building, interview prep, career counseling. We've worked with over 2,000 people getting them job-ready.

Both companies are solving the same core problem from different angles—the massive inefficiency in how talent and opportunity connect. Companies struggle to find the right people. People struggle to present themselves effectively. We're the bridge.

Everyone's talking about AI replacing recruiters. You're IN recruitment. What's actually happening?

AI is changing things, but not in the way the headlines suggest.

The tedious stuff—initial resume screening, scheduling interviews, basic qualification checks—that's getting automated. And honestly, that's great. Nobody wants to manually sort through 500 resumes.

But here's what AI can't do: It can't tell if someone's going to thrive in your company culture. It can't read between the lines of a resume to spot potential. It can't have the nuanced conversation that reveals whether someone's genuinely passionate about the role or just looking for any job.

We use AI heavily in our process. But it handles the grunt work so our team can focus on the human judgment parts that actually matter.

Hot take: The companies trying to fully automate recruiting with AI are going to end up with technically qualified people who are terrible cultural fits. You can't algorithm your way out of needing human judgment.

What's your take on the whole "ATS-friendly resume" industry?

Most of it is outdated advice that was maybe relevant five years ago.

People obsess over formatting, over specific keywords, over these arcane rules about how to structure a resume. They're optimizing for systems that don't work the way they think they do.

Here's what actually matters: Relevance to the job description. Can someone quickly understand what you've done and how it applies to what they need? That's it.

The best resume is one that tells a clear story about your progression and achievements. Not one that's gamed to trick an algorithm.

You've worked with 2,000+ people on their careers. What's the most common mistake you see?

Chasing opportunities without building depth.

People see a hot new field—data science, AI, blockchain, whatever—and they immediately pivot. They do a bootcamp, update their resume, and start applying.

But they haven't built real expertise. They're competing with thousands of others who did the exact same thing.

The people who succeed are the ones who go deep in something. They don't just learn Python; they become experts in applying Python to solve specific problems in a specific domain. They build a portfolio of real work, not just completed courses.

Depth beats breadth almost every time.

Hot take: Career advice that tells you to "learn to code" without context is useless. What matters is becoming genuinely valuable at solving a specific type of problem, not collecting certificates.

What about for people who are actively job hunting? What's your advice?

Stop treating job hunting like you're throwing darts blindfolded.

Most people apply to 100 jobs and hope something sticks. That's exhausting and ineffective.

Instead, target 10-15 companies where you actually want to work. Research them properly. Understand their challenges. Tailor your application to show how you specifically can help them solve their specific problems.

Then use your network. Find someone who works there. Get a warm intro. The vast majority of good hires happen through referrals, not cold applications through a job portal.

It's more work upfront, but your success rate goes up dramatically.

You bootstrapped both companies. What made you choose that path instead of raising funding?

Control and clarity.

When you bootstrap, you know exactly what your unit economics are. You know if you're actually building something sustainable or just burning through someone else's money.

It's harder. You grow slower. But you also build a real business instead of a venture-backed gamble.

Plus, I've seen too many founders raise money, then spend the next two years trying to hit metrics that their investors care about but that don't actually build a great product or serve customers.

What's been the hardest part of growing these companies?

Learning to delegate.

When you start, you do everything. You're the salesperson, the product person, the customer support, everything. And honestly, that works for a while.

But there's a point where your inability to let go becomes the bottleneck. I struggled with that. I wanted everything done exactly my way.

Learning to trust people, to build systems instead of doing everything myself—that was the unlock for scaling past a certain point.

Hot take: Most founders fail not because they can't build a good product, but because they can't build a good company. Those are different skills.

What's your view on the future of work? Everyone's predicting AI will disrupt everything.

AI will change what we do, but it won't eliminate the need for human workers.

Here's what I think happens: The boring, repetitive parts of most jobs get automated. The parts that require creativity, judgment, empathy—those become more valuable.

The people who thrive will be the ones who learn to use AI as a tool to amplify their capabilities, not compete with it.

In recruiting specifically, I think we'll see way more emphasis on soft skills, on cultural fit, on potential rather than just checking boxes of requirements. Because the technical screening part is getting easier, the human elements become more important.

You're in your 20s running two companies with dozens of employees. What do you wish someone had told you when you started?

Sales fixes everything.

You can have the best product in the world, but if you can't sell it, you have nothing. And I don't mean "sales" in the sleazy, pushy way. I mean genuinely understanding your customer's problems and communicating how you solve them.

I was technical. I was comfortable building things. Sales felt uncomfortable, even dirty.

But learning to sell, learning to have real conversations with potential clients, learning to close deals—that's what turned these companies from side projects into real businesses.

If I could go back, I would've focused on sales skills from day one instead of treating it as something I'd "figure out later."

Final Thoughts

Priyanshu's story isn't about ignoring AI or pretending technology isn't changing industries. It's about understanding what AI actually does well and building around that instead of against it.

While everyone was predicting doom for recruiters, he was busy using AI to make recruiting better. While others were talking about disruption, he was building two companies that serve thousands of clients.

The pattern that emerges from his journey is clear: Calculated risks beat blind leaps. Deep expertise beats surface-level knowledge. Sales skills matter as much as product skills. And technology is a tool, not a replacement for human judgment.

For anyone building in a space that's supposedly "threatened" by AI, Priyanshu's approach offers a roadmap: Use the technology, don't fight it. Focus on the parts humans do better. And build something that actually solves problems instead of chasing hype.

Because at the end of the day, AI might change how we work, but it doesn't change that businesses need talented people and talented people need good opportunities.

The companies that figure out how to connect those two things more effectively? They'll be around for a long time, AI or not.

Want more conversations like this?

This is what we do at ChaiNet—real conversations with people building real things. No rehearsed pitches, no corporate BS, just honest insights from founders, engineers, and operators in the trenches.

New episodes drop every week. Subscribe so you don't miss them.