Rose AI Search
Describe who you're looking for in natural language and let Rose research, source, and package real candidates for you.
Rose AI search is the fastest way to find candidates. Rather than setting filters manually, you simply describe who you're looking for in natural language — the same way you'd brief a recruiter — and Rose does the research for you.
Starting a search
Open the Rose home screen from the left sidebar. You'll be greeted with a prompt like “Hi, how can I help?” and a message box. If you've set up your Knowledge Base, you'll also see your company context shown above the box, with an Edit link.
Type a description of the role and candidate you're looking for. Be as specific or as broad as you like:
“I'm looking for an Account Executive with 3–5 years of experience at a NY tech startup”
“Find me a senior DevOps engineer based in London with AWS and Kubernetes experience”
A few helpful touches on the home screen:
- Upload a job description to have Rose read it and build the search for you.
- Use the sequence picker (Start a new sequence) to choose whether the candidates you find go into a new sequence or an existing draft.
- Chat history lets you reopen any past search and pick up where you left off.
What Rose does behind the scenes
When you send a brief, Rose works it in steps rather than firing off a single search. As she works, you'll see her progress — mapping out profile types, looking up companies that fit, finding candidates, and highlighting key profile details.
Clarifies the brief
If anything is ambiguous, Rose asks a focused question — often with tappable suggestions — or proposes a few candidate archetypes for you to choose between.
Finds the right companies
Rose identifies the companies your ideal hire works at today — competitors, adjacent players, and alumni networks — to use as high-signal starting points.
Sizes the pool first
Rose previews a search angle before committing, so she can refine the approach before retrieving profiles.
Retrieves and judges profiles
Rose pulls real profiles and runs a quick judgement pass against your must-haves, noting evidence and gaps for each candidate.
Understanding Candidate Packages
Rose returns her findings as Candidate Packages — curated groups of candidates that match a particular interpretation of your brief. Each package card shows how many candidates were found and a short label describing the angle (e.g. “Engineering managers with team-scaling experience”). Rose often returns more than one package so you can compare directions.
Click a package to open the full review:
- Candidates are ranked, deduped, and ready to review — the first 25 open straight away.
- Each candidate shows their photo, name, current title and company, relevant years of experience, and key skill tags.
- Add or Pass on candidates individually, or select several at once.
- Use Load next 25 to bring in more from the same package when you need extra volume.
- When you're happy, use Add [n] to send your selection to a sequence.
The experience number is smarter than it looks. Rose recomputes each candidate's relevant years of experience — counting only post-graduation, in-function work and excluding internships, student jobs, and unrelated roles.
Refining results in conversation
Because the search lives in a chat, you can steer Rose the way you would a colleague — ask her to narrow by location, seniority, or specific companies; tell her what you liked and didn't like; ask her to cap candidates per company or set a minimum experience bar; or ask her to reveal more from a package's reserve.
Rose is AI and can make mistakes. Always review profiles before adding them to a sequence.