Building GenAI Powered Mentorship Matching Tool
AWS She Builds Mentorship Program Team built an Generative AI tool that optimizes mentor-mentee matching leveraging AWS serverless services.
Lydia Ray
Amazon Employee
Published Jun 10, 2025
AWS SheBuilds Mentorship Program: Free 12-week program matching external mentees who identify as women with AWS mentors to help them enter or progress in technology careers.
The AWS She Builds Mentorship Program team has developed an innovative solution to enhance the mentorship matching process using generative AI. With huge number of applicants and varying preferences, creating quality mentor-mentee matches is critical to program success but has become a bottleneck as the program scales globally.
In this session, the team showcases how they built an AI-powered matchmaking tool that streamlines the matching process without requiring complex infrastructure management. The system combines serverless AWS services with responsible AI principles to analyze applications, identify compatibility factors, and generate optimized mentor-mentee pairings, while still maintaining human oversight for final matching decisions. This approach dramatically reduces the manual effort required while improving match quality across multiple dimensions.
- Problem Solved: The tool addresses the challenge of manually matching huge number of mentees with mentors across global regions, which had become a significant bottleneck to program scaling.
- Technical Architecture: Built using serverless AWS services including AWS Lambda, Amazon S3, Amazon EventBridge, AWS Step Functions, Amazon DynamoDB, and Amazon Bedrock for GenAI capabilities.
- Responsible AI Integration: Implements multiple dimensions of responsible AI including fairness (removing PII), explainability (providing reasoning for matches), safety (using Bedrock guardrails), and transparency.
- Human-in-the-Loop Design: AI generates recommendations but program managers make final decisions, maintaining the critical human element in mentorship matching.
- Development Approach: The team emphasized starting small but thinking big, and assembled a diverse development team to ensure multiple perspectives informed the tool design.
- Evolution Plans: Future improvements include adding a proper UI, incorporating user feedback from the hybrid approach, and potentially adding agentic capabilities.
Hosts of the show 🎤
- Sr Analytics Solutions Architect @ AWS
- Global Program Manager @ AWS She Builds Mentorship Program
Guest 🎤
- Partner Solutions Architect @ AWS
- Global Program Manager @ AWS She Builds Mentorship Program
- Sr Specialist Solutions Architect, Amazon Q Developer @ AWS
- GenAI Builder Team @ AWS She Builds Mentorship Program
- Sr Prototyping Architect @ AWS
- GenAI Builder Team @ AWS She Builds Mentorship Program
Any opinions in this post are those of the individual author and may not reflect the opinions of AWS.