How UX and AI teamed up to transform recruiting at Amazon
Summary
I led UX research and product design to evolve an AI chatbot used by Amazon recruiters, addressing knowledge gaps, building features to support learning through human collaboration, and ultimately improving engagement, task success, and productivity. My work helped recruiters save time, share knowledge securely, and attract top talent, while reducing reliance on costly third-party tools.
35%
Improved onboarding process
35%
Improved onboarding process
How UX and AI teamed up to transform recruiting at Amazon
I led UX research and product design to evolve an AI chatbot used by Amazon recruiters, addressing knowledge gaps, building features to support learning through human collaboration, and ultimately improving engagement, task success, and productivity. My work helped recruiters save time, share knowledge securely, and attract top talent, while reducing reliance on costly third-party tools.
35%
Improved onboarding process
25%
Increase in user retention
84%
Increase in time spent on website
What I Do
Amazon employs 5,000+ recruiters drowning in a dozen disconnected tools – from Slack to internal Wikis and emails. Manual knowledge hunting wasted hours daily.
Amazon employs thousands of recruiters across teams and time zones—each juggling tools like email, Slack, internal wikis, and Quip to find or share critical information. Fragmentation led to inefficiencies and lost knowledge, especially when switching contexts or onboarding new recruiters.
Full-stack designer crafting digital experiences that will make you product or service in awesome way.
Problem / Opportunity & Goals
Problem: TILT was only as smart as the knowledge it indexed—which meant it often couldn’t help with specific or uncommon queries. TILT’s AI could only answer 30% of recruiter questions – leaving critical gaps in live hiring scenarios. Recruiters needed answers, fast. Many turned to Slack, where informal Q&A, resume sharing, and discussions happened—but on an external tool with privacy, security, and cost concerns.
We saw an opportunity:
Could we enhance TILT with human-powered knowledge sharing—making it both more helpful now and smarter over time? We could turn unanswered questions into crowd-sourced wins while training the AI.
Success Metric(s):
Knowledge Coverage: Resolve 90% of recruiter queries
Improve task completion rates with TILT
Increase user trust and engagement with AI
Security: Reduce Slack reliance and resume sharing outside internal tools
Create a scalable, secure system for recruiter knowledge exchange
Cost: Replace $1M+ Slack dependency
Challenges
Stakeholder Skepticism: Engineering saw community features as "scope creep"
Tool Fragmentation: Recruiters refused another platform
Compliance Nightmares: Sharing candidate data internally required legal firewalls
Team, Role, & Responsibilities
I joined as a UI/UX designer shortly after the chatbot beta launched, embedded on a cross-functional team that grew from 7 to 14. My responsibilities included:
Leading research to identify user pain points and opportunities
Partnering with data engineering on discovery and validation
Designing core features: chat interfaces, feedback loops, human Q&A
Contributing components to AWS’s design system
Supporting handoff and quality assurance with engineering
Collaboration & Alignment: Cross-team alignment was essential. I worked closely with PMs, AI researchers, engineering, and recruiting leaders to define success and build trust. A key part of my role was navigating competing priorities—balancing chatbot limitations, security concerns, and recruiter urgency—to co-create solutions that worked for everyone.
“ With our new visual branding and language in place, the new Shopify brand clearly captures the essence of our current and target customer base, our employees, and our values. ”
Tobias Lütke
CEO, Co-founder | Shopify
Conclusion
The modernization of the subscription management platform successfully addressed the core usability issues and improved the overall user experience. By focusing on simplifying the interface and optimizing workflows, we were able to create a more efficient and enjoyable platform for users. The significant improvements in user engagement, satisfaction, and subscription rates underscore the importance of user-centric design in achieving business success.
How UX and AI teamed up to transform recruiting at Amazon
I led UX research and product design to evolve an AI chatbot used by Amazon recruiters, addressing knowledge gaps, building features to support learning through human collaboration, and ultimately improving engagement, task success, and productivity. My work helped recruiters save time, share knowledge securely, and attract top talent, while reducing reliance on costly third-party tools.
35%
Improved onboarding process
25%
Increase in user retention
84%
Increase in time spent on website