Intelligent Job Recommendation System

Problem It Addresses

Job seekers often face fragmented job descriptions, outdated salary data, and hard-to-spot role risks. Recent graduates, career switchers, people with long employment gaps, and aspiring founders need to weigh skill fit, interests, city opportunities, family responsibilities, and long-term growth at the same time. This skill turns job recommendation into a repeatable multi-dimensional evaluation workflow.

How It Works

• Information gathering: collects education, skills, salary expectations, location preferences, company culture preferences, family context, and entrepreneurship intent through dialogue.
• Real-time search: calls web_search before recommending roles to gather current job requirements, market salary, and industry trends.
• Multi-perspective matching: combines HR, recruiter, career coach, psychologist, city advisor, and ten other perspectives, with weights adjusted by user type.
• Structured output: provides matching score, expert assessment, role details, job-seeking strategy, alternative plans, and safety baseline notes.
• Quality checks: recommendation_quality_check.py can verify safety checks, emotional support, real-time search, and coverage of expert perspectives.

Boundaries

• Useful for direction, role comparison, and interview preparation, not a replacement for real hiring platforms or employer decisions.
• If user information is too sparse, key context should be added before matching.
• Output emphasizes legal and legitimate roles and avoids black/gray markets, but cannot guarantee the final truthfulness of every listing.
• Long-gap and switching cases include emotional support, but recommendations should not be read as hiring guarantees.