Designing AI for Real Decisions: From Connections to Opportunities

Disclaimer: Future World Alliance welcomes guest speakers to provide education and diverse perspectives. Any views, opinions, or statements made by the guest speaker are solely their own and may not represent the views of Future World Alliance. ​​Session Description: ​In a world full of data and connections, making the right decision remains the hardest part. This session explores how AI can be designed not just to generate outputs, but to support meaningful decision-making. ​Using a real system I built that helps users identify high-quality opportunities from their professional connections, I will walk through how to structure data, design ranking logic, and selectively use AI to enhance clarity rather than replace reasoning. The focus will be on building systems that combine deterministic analytics with AI-generated insights to produce results that are both useful and interpretable. ​Participants will gain practical insights into designing AI systems that go beyond chat interfaces—systems that prioritize relevance, transparency, and user trust. I will also share key lessons from building and iterating on this system, including challenges with data quality, evaluating AI outputs, and balancing automation with human judgment. ​This session is designed for educators, innovators, and creators looking to build AI-driven tools that are not only intelligent, but genuinely helpful in real-world decision contexts. ​Session lead: Samridhi Vats ​I am analytics professional at Amazon, where I work on building data systems that translate complex data into real-world decisions. My work focuses on how insights are generated, validated, and ultimately used to drive impact at scale. ​I hold a Master’s degree in Business Analytics from Purdue University and am a Certified Analytics Professional (CAP) through INFORMS. Beyond my role, I actively engage with the analytics community through conference presentations, workshops, and mentoring early-career professionals. ​I am particularly interested in moving beyond surface-level AI applications to designing systems that are reliable, interpretable, and genuinely useful. My work reflects a broader goal: bridging analytical thinking with practical, real-world problem solving.