CloudHire: From Job Description to Candidate Recommendations

Most sourcing tools put the search burden on the recruiter to write the query, set the filters, scroll the results, repeat. CloudHire’s recommendation feature flips part of that process. Instead of starting with a blank search box, recruiters start with the job itself, and the platform surfaces candidates from its internal pool based on what the role actually requires.

Here’s how the feature works, step by step.

CloudHire AI Job Recommendation Feature:

Starting With the Job Description

When a recruiter creates a job, they have a few options for how to define it. They can upload an existing job description, write a short search query in plain language, or enter the key details manually. If they write a basic search query, the AI analyzes it and generates a full job description from it, including core skills, relevant titles, and role context. This works well when a recruiter has a clear sense of what they need but hasn’t written a formal JD yet.

After the job description is in place, the recruiter selects the experience level and budget range they’re targeting. These become part of the recommendation logic, candidates outside the salary range or experience band won’t appear in the results.

Reviewing and Adjusting Recommendation Parameters

Before candidates are shown, the platform moves to a review section where the recruiter can see the parameters that will govern the recommendations: Job titles, required skills, optional skills, location, budget range, and experience level. Each of these can be adjusted.

For titles, the platform suggests additional options that might expand the search. For skills, there’s a split between required and optional. Recruiters can move skills between these categories, remove ones that aren’t relevant, or add new ones. The location can be set manually, or from AI suggestions, the budget and experience sliders can be refined before any candidates are shown.

This review step matters because it gives the recruiter control over what ‘best match’ actually means for their specific role, rather than accepting the AI’s interpretation as-is.

What the Candidate Results Look Like

Once the job is created, CloudHire sources candidates from its internal Cloud ID pool and populates the Cloud ID Verified section. Each candidate card includes their AI interview score, years of experience, relevant skills, location, and current and expected salary. Recruiters can see at a glance how a candidate aligns with the role’s parameters without having to open each profile individually.

Clicking through to a candidate profile shows a fuller picture: their profile summary, uploaded resume, interview recordings, and interview analysis. There’s also a chat option where recruiters can start a conversation with a candidate directly from the platform. Resumes can be downloaded, and candidates can be shortlisted with a single click, which moves them into the screening pipeline.

Updating the Search After Results Come In

If the initial results aren’t quite right, say the recruiter wants to narrow by location, or the experience range needs adjusting, the search parameters can be updated without starting over. The recruiter changes the filters and clicks ‘Save and Regenerate,’ and the platform shows a fresh set of candidates based on the updated criteria. The previous results will be updated with the new filtered set.

This iterative approach means the recruiter doesn’t have to get the parameters perfect on the first pass. It’s closer to how search actually works in practice, you see results, refine, and get closer to what you need.

How This Fits Into the Broader Hiring Flow

The recommendation feature handles candidate discovery for jobs posted through CloudHire. From there, shortlisted candidates move into the hiring pipeline, where recruiters can manage stages, send templated communications, run AI interviews, and schedule through Google Calendar. The other posts in this series cover each of those parts of the process.