Recruiters Use LLM Screening Chatbots to Filter Portfolios — Designers Need New Metadata

AI · 4 min read

To manage high volumes of applications, many recruiting teams have implemented LLM-driven screeners that extract structured data from portfolios: metrics, role scope, tools used, and AI workflow notes.

Designers whose portfolios are purely visual risk being deprioritized because these systems favor explicit, machine-parsable evidence of impact. Adding short metadata tags, structured case study fields, and JSON-style project summaries helps pass automated screens.

However, teams still include human review for finalists, so designers must balance machine-friendly structure with compelling narratives and artifacts that demonstrate authorship.

Practical steps include adding a one-paragraph machine-summary per case study, including bulletized outcomes, and appending a short 'AI used' field where relevant — small changes that improve discoverability in recruiter workflows.