Expert-level screening, around the clock.

Duke Talent is a specialist recruiting firm serving the architecture and interior design industry — a world where the right placement comes down to things a job board can't see: design sensibility, project type, studio culture, career arc. Duke Talent's edge has always been judgment: knowing which designer belongs at which studio, and having the network to make the introduction.

The problem with that kind of judgment is that it doesn't scale. Every hour the team spent screening resumes, cross-referencing spreadsheets, or combing job boards was an hour not spent with designers and studios. So rather than simply hiring its way out of the bottleneck, Duke Talent partnered with Linkt to encode its expertise into intelligent systems and scale the judgment itself.

Duke Talent's matching criteria were sharp. The team could tell quickly whether a candidate fit a studio. But applying that judgment meant manual work across a scattered landscape: candidate and company records in the firm's existing marketplace, LinkedIn profiles and resumes in every imaginable format, portfolio PDFs, job listings spread across industry boards, and spreadsheets, notes, and institutional memory.

Every candidate needed a manual first pass. Every search started from scratch. The knowledge that made Duke Talent valuable lived in its recruiters' heads, and the day only has so many hours.

Linkt didn't arrive with an off-the-shelf product to configure. Working forward-deployed alongside Duke Talent's team, Linkt's engineers captured how the firm actually evaluates candidates, ranks studios, and decides what a great match looks like — then built that thinking into three connected systems.

The intelligence underneath is Duke Talent's, formalized: candidate scoring tuned to the firm's standards, company classification that ranks studios the way its recruiters would, and plain-English fit summaries that explain each match like a recruiter would. To ensure fairness, every candidate is reviewed against the same criteria, applied the same way. And all of it plugs directly into Duke Talent's existing marketplace: members click through and are signed in automatically. No new logins, no parallel systems, no friction for the community the firm had already built.

A firm's judgment doesn't fit in a requirements document. What began as a matching prototype grew into a full product suite, shaped throughout by the people using it: every major capability started with a written design spec, direct input from Duke Talent's team set the priorities, and when a better approach emerged, Linkt rebuilt. The original graph-based matching engine, for example, gave way to a faster search-and-ranking system as the platform matured.

The result behaves less like a vendor deliverable and more like a product built by an embedded team — because it was.

Work that used to consume the day now runs in the background. Every candidate gets a thorough, expert-level first review in minutes, including the ones who apply at midnight. Recruiters open a role and start from a ranked shortlist instead of a blank search bar. Every expression of candidate interest lands in an actionable queue, with the person, the role, and a fit summary side by side. And the data stays current on its own.

Duke Talent's judgment now shows up in every search, every review, and every match, while the team's time goes to the part of the job software can't do: relationships with designers and studios.

Generic recruiting software is built for the average staffing firm, and averages are exactly what Duke Talent isn't. A specialist firm's edge is its taste, criteria, and network: the hardest thing to put into software, and the most valuable.

That's what Linkt's forward-deployed model is for — engineers embedded with the client's team, turning how its experts actually work into production systems, and refining them continuously as the business evolves.