Silent Infrastructure: How Privacy-First SDKs Quietly Shape the Future of Global-Scale Mobile Apps
Tuesday, August 4, 2026 at 2:35 PM to 2:55 PM · 20 min. (US/Pacific)
Lando Ballroom 4204
Solo Talk
Software Engineering
Information
What if the most critical layer of your mobile infrastructure is also the most invisible?
Mobile Software Developer Kits (SDKs) silently power billions of experiences every day—from AI-driven recommendations to real-time payments, from identity verification to on-device machine learning. Yet despite their ubiquity, SDKs remain the most underestimated component of the mobile ecosystem. When designed well, they're invisible. When designed poorly, they become front-page news—crashing apps, leaking data, and eroding user trust at scale.
In this session, I'll pull back the curtain on the engineering principles that separate world-class SDKs from technical liabilities. Drawing from over 12 years of building mobile infrastructure at LinkedIn, Google Firebase, Snap, Open Garden, and Samsung — serving billions of users and millions of developers—I'll share hard-won lessons on architecting SDKs that are privacy-first, globally scalable, and genuinely developer-friendly.
Why does this matter now more than ever? As AI capabilities are increasing, mobile SDKs are becoming the delivery mechanism for sophisticated ML models, privacy-preserving computation, and intelligent features that users interact with daily. The decisions we make at the SDK layer—how we handle data, how we design APIs, how we balance performance with compliance—directly shape what's possible for AI on mobile. Get it wrong, and you've built a foundation of sand. Get it right, and you've created infrastructure that quietly enables innovation for years to come.
The future of mobile AI isn't just about models—it's about the invisible systems that deliver them. Join me to explore how thoughtful SDK design doesn't just enable mobile apps—it quietly defines their future.
Mobile Software Developer Kits (SDKs) silently power billions of experiences every day—from AI-driven recommendations to real-time payments, from identity verification to on-device machine learning. Yet despite their ubiquity, SDKs remain the most underestimated component of the mobile ecosystem. When designed well, they're invisible. When designed poorly, they become front-page news—crashing apps, leaking data, and eroding user trust at scale.
In this session, I'll pull back the curtain on the engineering principles that separate world-class SDKs from technical liabilities. Drawing from over 12 years of building mobile infrastructure at LinkedIn, Google Firebase, Snap, Open Garden, and Samsung — serving billions of users and millions of developers—I'll share hard-won lessons on architecting SDKs that are privacy-first, globally scalable, and genuinely developer-friendly.
Why does this matter now more than ever? As AI capabilities are increasing, mobile SDKs are becoming the delivery mechanism for sophisticated ML models, privacy-preserving computation, and intelligent features that users interact with daily. The decisions we make at the SDK layer—how we handle data, how we design APIs, how we balance performance with compliance—directly shape what's possible for AI on mobile. Get it wrong, and you've built a foundation of sand. Get it right, and you've created infrastructure that quietly enables innovation for years to come.
The future of mobile AI isn't just about models—it's about the invisible systems that deliver them. Join me to explore how thoughtful SDK design doesn't just enable mobile apps—it quietly defines their future.
Job Function Tracks
Software Engineering
