Capability Gap

All companies

Juicebox

PeopleGPT: semantic search over 800M+ profiles at $99–179 a month. The tooling end of the recruiting market, and the price floor under everything in it.

medium confidence4 minupdated 2026-08-29recruiting · sourcing · software · price floor
Vertical
Recruiting
Founded
2022 (YC S22)
Headquarters
Not in the record
Raised
$36M
Last valuation
Never disclosed
Revenue
$10M+ ARR at Series A
Status
Active

Juicebox sells software, not placements, and that is exactly why it belongs in the Contingency recruiting marketplaces vertical rather than beside it. A $179/month seat that lets a founder describe the person they want in a sentence and search 800M+ profiles for them is a direct substitute for a $58,000 placement fee — if the founder's time is free. It is the same outcome priced at roughly 1/300th of the cost, and it sets the floor the whole fee ladder has to sit above.

What is known

  • YC S22. Founders David Paffenholz and Ishan Gupta, 22 and 19 at founding.
  • $30M Series A led by Sequoia (David Cahn); $36M total (TechFundingNews; PeopleMatters). No valuation disclosed. [UNVERIFIED]
  • $10M+ ARR and 2,500+ customers at the Series A, with no formal sales team — pure word of mouth (TechFundingNews). 12 employees (Sacra).
  • Now claims 5,000+ customers and 800M+ searchable profiles (Juicebox) — a company-authored source, so treat the customer count as a company claim rather than an audited figure.
  • Pricing $99–$179/month, with a free tier (HeroHunt).
  • Logos: Cognition, Ramp, Perplexity.
The numbers

$10M ARR across 2,500 customers is a ~$4,000 average contract value — consistent with the published $99–179/month seat prices plus multi-seat accounts. Twelve employees against $10M+ of software revenue is roughly $830K per head, which is a genuine SaaS profile and not a services one. This is the only company in the vertical whose revenue is unambiguously net revenue; nothing has to be backed out of a gross number. See GMV is not revenue.

Why it matters strategically

Juicebox is the disintermediation threat to the marketplace model, priced so far below it that the comparison barely registers as competitive. The explicit pitch is that founders have "hired entire teams using Juicebox without external recruiters" (TechFundingNews). Set against the published cost-per-hire ladder — agency contingency at 20% of a $100K role = $20,000, fractional recruiter ≈ $8,000, in-house average $5,475, AI sourcing platform ≈ $1,200/year (Pin) — the spread between the top and the bottom of that ladder is the entire strategic tension in the market.

There is one strong argument on Juicebox's side that has nothing to do with price. 71% of AI/ML roles are filled by candidates whose current title is not "AI" or "ML" — they are backend, infra or research data scientists (Pin). Title-and-keyword search structurally misses them. Semantic search is the correct tool for that problem, and the boolean-search agency is the wrong one.

The counter-argument

Sourcing was probably never the binding constraint. Finding the candidate is the cheap half; persuading a happy senior engineer at Anthropic to take a call, calibrating a bar with a founder who has never hired the role, and closing an offer against three competing bids are the expensive half, and none of them are search problems. A 15–25% contingency fill rate (Bullhorn GRID 2026 via Pin) is not a statement about how hard it is to find people.

The other counter-argument is that the founder's time is not free, and a 37% rise in time-to-fill against 244 applications per open role (TheHireHub) is precisely the condition under which a founder stops doing their own sourcing.

So what

Both things can be true, and the resolution is a segmentation rather than a winner. Juicebox takes the middle of the role distribution — the roles where finding the candidate is most of the work. Paraform keeps the scarce tail, where its $260K average placed comp says it already lives. The commodity agency in between is the one that dies, and it is dying now.

What to watch

Whether Juicebox stays a tool or moves toward outcomes. Sequoia's own thesis — the one that names Juicebox — argues that as models cross a competency threshold, the companies that sell the work beat the companies that sell the tool (Sequoia). A tool vendor sitting on 5,000 hiring companies and 800M profiles has an obvious path into placement fees, and every incentive in its investor's published thesis pushes it that way. If it takes that path it stops being the price floor under the vertical and becomes a competitor inside it — see Marketplace, staffing firm, BPO or agency and What the model actually is.

Also worth noting for anyone modelling the category: at 12 employees and $10M+ ARR with no sales team, Juicebox has the best capital efficiency in the entire recruiting set, and it got there with $36M it visibly did not need. Toptal made the same point a decade earlier with $1.4M.