On a Thursday morning in mid-March, a three-person startup from Y Combinator went live on Product Hunt with a bold proposition: hand over your entire recruiting operation to an AI agent named Paul.
No human recruiters. No email chains with agencies. Just an autonomous system that sources candidates, runs preliminary screens, and delivers interview-ready profiles straight into Slack—all within 24 hours of kicking off a search, or so the pitch goes.
Perfectly, the startup behind Paul, emerged from Y Combinator's Winter 2025 batch with the kind of timing that's either shrewd or accidental. The very same day they launched—March 13—Workable, a well-established player in applicant tracking systems, unveiled its own embedded recruiting agent. The message was hard to miss: the race to automate hiring is no longer theoretical.
What's striking isn't just that these tools exist. It's how quickly the landscape has filled in around them.
A Crowded, Moving Target
In August 2025, Eightfold AI rolled out a "Recruiter Agent" aimed at high-volume hiring. hireEZ pitched its "Agentic AI" approach, adding resume integrity checks by December. hackajob introduced three AI voice agents in October to strip away administrative drag. And Workday, never one to be left behind, is in the process of acquiring Paradox, the company behind Olivia, a conversational AI platform that's been in the wild for a while now.
Gartner weighed in with a January prediction: by 2028, 60 percent of brands will deploy agentic AI for one-to-one customer interactions. A February report from the Josh Bersin Company described "AI-powered superagents" as the catalyst for the largest HR transformation in decades. Recruiting agents, naturally, sat at the top of the list.
Perfectly is betting it can cut through that noise. The founders—Victor Luo, Zhuang (Gary) Luo, and Huimin Xie—are ex-TikTok machine learning engineers, with Gary also clocking time at Meta. They frame recruiting not as a traditional HR problem but as a recommendation-systems challenge, the same logic that decides which video you watch next or which ad follows you across the internet.
"We built systems that matched billions of users to content," the team wrote in their early February YC Launch post. "Matching candidates to roles is fundamentally the same problem."
That framing shows up in how Paul actually operates.
What Paul Does (and What It Claims to Do)
The workflow starts with an intake session—modeled, the founders say, on the recommendation pipelines they built at TikTok and Meta. From there, Paul begins sourcing candidates, sending outreach messages, and conducting initial screens. Interview-ready profiles appear in Slack. A hiring manager reviews, clicks approve, and Paul fires off a calendar link for the first real conversation.
Perfectly claims it can fill a manager's interview calendar within a day and close most roles in two to four weeks. The company reports a tenfold jump in interview volume during the first week after onboarding, roles filled four times faster than traditional methods, and candidates passing interviews at twice the typical rate.
Those numbers come directly from Perfectly's YC profile and website, captured in March, and represent the company's own self-reported metrics. No independent verification exists yet. No tier-one tech publication has covered the launch in depth. What's out there is mostly YC amplification, Product Hunt chatter, and a handful of blog write-ups.
The pitch leans heavily on continuous calibration. Paul has "infinite memory and context," according to the company, adjusting in real time as hiring managers accept or reject candidates. The idea is that the agent gets smarter with every interaction, fine-tuning its understanding of what a given manager actually wants—not just what the job description says.
Whether that happens in practice is, for now, an open question.
Undercutting Agencies (in Theory)
Perfectly charges a success-based agency fee, typically 15 to 25 percent of first-year salary. The company positions this as roughly half the cost of traditional recruiting firms—a claim that holds up if you're comparing against the high end of agency pricing. The YC Launch offer sweetened the deal with an additional 50 percent discount on the first role for anyone mentioning the announcement.
There's also Parker, a candidate-facing product launched the same day. Parker markets itself as an "AI career super-connector" on iMessage and WhatsApp, designed to get job seekers in front of hiring managers faster. Both products went live on Product Hunt simultaneously, with the main Perfectly agent landing fifth for the day—a respectable showing, though not a runaway win.
The business model is familiar to anyone who's worked with recruiting agencies. The difference, Perfectly argues, is speed and volume. No human recruiters means no ghosting on difficult searches, no bottlenecks between candidate qualification and hiring manager review, and no need to log into yet another portal.
Everything lives in Slack, where most startup teams already spend half their workday anyway.
The Founders' Origin Story
The founding narrative is straightforward, almost generic in the way early-stage startup stories often are. The team got frustrated with legacy recruiting agencies—slow pipelines, opaque processes, roles that dragged on for months. They looked at the problem through the lens of machine learning and decided they could build something better.
In fairness, they probably can build something. Whether it's better is another matter.
Victor, Gary, and Huimin spent their careers optimizing engagement and personalization at massive scale. TikTok's recommendation engine is arguably the most addictive product of the past decade. Meta's ad targeting system is a $100 billion-plus business. Both rely on the kind of iterative learning and pattern recognition that Perfectly is now applying to candidate matching.
But people aren't videos. Hiring isn't content discovery. The stakes are higher, the feedback loops messier, and the margin for error slimmer. A bad recommendation on TikTok costs you 15 seconds. A bad hire costs a company months and tens of thousands of dollars, sometimes more.
That's the tension Perfectly—and every other recruiting agent startup—has to navigate.
Reality Check
The claims deserve scrutiny. A 10× increase in interview volume sounds impressive until you consider the baseline. If a company was scheduling two interviews a week and suddenly schedules twenty, that's transformative. If they were already running a high-volume pipeline, the marginal gain might be less dramatic. Perfectly hasn't disclosed customer names, case studies, or verifiable data beyond what's on its own website.
The 2× pass rate—candidates advancing through interviews at double the normal rate—is particularly intriguing, if true. It suggests Paul is either exceptionally good at pre-screening or that hiring managers are lowering their standards to fill calendars faster. The company would argue it's the former. Skeptics might wonder about the latter.
And then there's the broader question of whether fully autonomous recruiting agents actually work at scale. Industry commentary from early 2026—admittedly scattered and anecdotal—points to mixed results. Some companies report faster pipelines and better candidate quality. Others note that human oversight remains essential, especially for senior or highly specialized roles where cultural fit and nuanced judgment matter as much as technical qualifications.
A Slack whitepaper from around the same time framed AI agents as "teammates" embedded in the platform, which aligns neatly with Perfectly's Slack-native delivery model. But the metaphor only holds if the agent actually behaves like a teammate—reliable, adaptable, and capable of handling ambiguity. If it behaves like a particularly aggressive spam filter, the whole system falls apart.
What Happens Next
Perfectly is targeting startups and growth-stage companies—organizations that move fast, operate lean, and are generally more willing to experiment with unproven tools. That's a sensible beachhead strategy. It's also a competitive one. Eightfold, hireEZ, hackajob, and now Workable are all fishing in the same pond.
The company lists three employees on its YC page as of March, though that number may have shifted as the product gained traction. No public funding details are available beyond the standard YC investment. The team hasn't disclosed revenue, customer count, or how many roles Paul has actually filled.
What Perfectly does have is momentum, or at least the appearance of it. A Product Hunt launch. YC backing. A clean narrative. And a product that addresses a real pain point: recruiting is slow, expensive, and often frustrating for everyone involved.
Whether Paul can deliver on the promise of same-day interview pipelines and two-week time-to-fill will depend on execution in a market where several well-funded competitors are making nearly identical bets. The next few months will clarify whether Perfectly is leading the pack or just another name in a very crowded field.
For now, the pitch is compelling. The proof, as always, will be in the hires.
