Skip to content

MCP onboarding: role suggestion calibration #17

@Mayank-glitch-cpu

Description

@Mayank-glitch-cpu

Tracking tradeoff from agentic MCP migration plan (onboarding flow).

Context: After extracting the user's resume, the agent proposes a detected role list saved to ~/.jobsync/profile/roles.json. User then confirms, adds custom roles, and marks excluded roles.

Open questions:

  • How aggressive should the agent be? Suggest top 3 roles vs top 10?
  • Weight by recency of experience (e.g., most recent job > projects from 3 years ago)?
  • Handle career-switchers: someone with backend experience targeting AI roles
  • Should custom roles be validated against a known role taxonomy or accepted verbatim?

Downstream impact: union of detected + custom minus excluded drives every scrape_jobs_workflow run — miscalibration means irrelevant jobs in Airtable.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions