grade-jobs

To automatically evaluate job listings, this Skill grades roles stored in a local SQLite database against your structured candidate profile and career goals. It processes pending queues by grounding assessments in fetched job descriptions and generating evidence-based fit scores.

The best way to filter a large backlog of job descriptions is to use lane-relative grading, which evaluates roles against peers within their specific functional lane. This autonomous pipeline prevents grade drift and highlights only the top matches aligned with your trajectory.

Yes, this job grading engine requires a local SQLite database for data persistence. You must store your pending job listings in SQLite to enable the multi-phase pipeline of parallel initial grading and within-lane relativity cross-checks.

Lane-relative grading evaluates job listings against peers within their specific functional lane, such as systems-infra or ai-ml, rather than using a global scale. This prevents inaccurate comparisons across disparate roles and ensures consistent, cross-checked fit assessments.

Yes, you can use this Skill to generate structured fit assessments that explicitly cite your specific technical projects, skills, and career goals as evidence. The engine grounds its evaluations in actual job description content to produce these prose-based justifications.

The automated job grading process includes a relativity pass to perform cross-check reviews across your database, preventing grade drift. This ensures that generated fit assessments and grades remain consistent and accurate relative to peer listings within the same lane.