Build intelligent mentor-mentee matching system with skill alignment and progress tracking

[anshul23102]

Description

HerStack emphasizes 'women mentors' but lacks automated matching mechanism. Current manual coordination is unscalable, cannot leverage mentor expertise or mentee goals. Results in poor pairings (40% mismatch rate), frustrated users, months-long wait times.

Current Impact: Manual matching breaks at scale. Mentors overwhelmed (100+ requests each). Mentees waiting 2-3 months for assignment. Program cannot grow beyond 50 mentors.

Expected Business Value: Automated matching reduces assignment time 90%, improves satisfaction 80%, enables scaling to 500+ mentors, increases mentor retention 70%, positions HerStack as professional mentorship platform.

Steps to Reproduce

1. Sign up as mentee seeking ML engineering mentor
2. Look for mentor search or matching system
3. Observe: no matching interface; must contact admin
4. No visibility into mentor backgrounds or expertise
5. Cannot view available mentors

Environment Information

• Frontend: React mentor discovery UI
• Backend: Node/Python matching API
• Database: mentor and mentee profiles
• ML: skill similarity algorithms
• Email/Notifications: For matching notifications

Expected Behavior

• Mentee completes profile: goals, skills, timezone, availability
• System analyzes profile and identifies compatible mentors
• Shows ranked mentor list with match scores (80%, 85%, etc.)
• Displays mentor profiles: expertise, experience, mentee feedback ratings
• Mentee selects preferred mentor; system initiates introduction
• Track mentorship: meetings logged, progress toward goals
• Collect feedback: mentee satisfaction, mentor feedback

Actual Behavior

• No mentor matching system
• Manual admin coordination
• Cannot see mentor qualifications
• Unscalable process

Additional Context

Affected Users: Mentees unable to find suitable mentors; mentors overwhelmed with requests.

Root Cause: No matching algorithm or mentor discovery system.

Proposed Solution: Skill-based matching with preference weighting and satisfaction tracking.

Implementation Steps:

1. Design mentor profile schema with expertise areas, experience, teaching style
2. Design mentee profile with learning goals, desired mentor skills
3. Implement skill-based matching: overlap_score = matched_skills / desired_skills
4. Add preference factors: timezone, industry, experience level, teaching style match
5. Weighted scoring: skills(0.4) + availability(0.3) + experience(0.2) + feedback(0.1)
6. Build mentor discovery UI: search, filter, view profiles
7. Implement mentorship agreement workflow
8. Create progress tracking: log meetings, goals, feedback
9. Build mentor health dashboard with engagement metrics
10. Create feedback collection and satisfaction surveys

Test Cases:

1. Exact skill match: mentee wanting Python+ML matched with mentor expert in both (expect >90% score)
2. Partial match: mentee wants 3 skills, mentor has 2 (expect 66% match, still recommended)
3. Timezone alignment: same timezone gets higher score than different timezone
4. Experience preference: mentee preferring 5+ years matches with experienced mentors first
5. Feedback influence: higher-rated mentors ranked above equal-skill mentors
6. Path shows top 3 options with clear explanation
7. Selected mentor notified, mentee notified, meeting scheduled
8. Progress meetings logged, tracked over time
9. Quarterly feedback collected and stored
10. Mentor burnout detected: 15+ mentees = warning to take break

Severity: High - critical for mentor program
Expected Points: 700-800 GSSoC points

Suggested Labels

enhancement, matching-algorithm, mentorship, community, GSSoC26