Michael Sayman

“You're never as good as they say when you're up, and you're never as bad as they say when you're down. You just have to stay focused and keep shipping over the long term.”

I’m excited to share that I’m joining @Apple to help build UI frameworks and animations! I submitted my first app to the App Store in 2010. Looking forward to working alongside those who made it all possible. 🍎

Shana Tova! 🍎🐝🍯 hoping for a sweet new year 💛

we’re married! 💍🥂

An incredible run for Tim Cook over the past 15 years! Apple has shaped my life in so many ways over these years, and I can’t wait to see what comes in this next era. 😊❤️

Me, my braces, and the first app I launched on the App Store in 2010.

When I was approached to work on creating a version of the Meta AI ring UI code specifically for our new Meta Ray-Ban Display glasses, I leaped in excitement. 😅 It’s been surreal working w/ the Wearables team on this over the past months. Grateful to be a part of this moment.

I’m excited to share what I’ve been working on at Meta. I was tasked with building + designing a reimagined Meta AI embodiment across all our platforms + services. It’s been so much fun working on delivering craft as best as I know it for Meta. I used Metal shaders in iOS, OpenGL in Android, and WebGL for web to build this centralized new embodiment. The new ring is now rolling out across Facebook, Instagram, WhatsApp, and Messenger on iOS, Android, and on Meta AI web.

information loss in English when transforming from verbal to written text iMessage should automatically add those "emphasis animations" on certain words in sentences when it detects their emphasis in an audio transcription it'll come soon enough

My design tool of choice over the years: 2000-2003: Crayons 2004-2007: KidPix 2008-2013: Keynote 2014-2022: Xcode 2023-present: AI Coding IDEs

10 years ago today was my 1st day as a full-time engineer at Facebook. Forever grateful for those years. ❤️

had fun jumping on The Verge’s podcast this week to explain why I made SocialAI

The SocialAI experiment has been so magical. For many years, I wanted to give this a shot, and I’m so glad I went for it and did. ❤️ I’d like to share 4 key things I’ve learned that could help others on their journeys building AI consumer products: 1. The Interface Shapes the Experience The interfaces we use for products significantly define how we perceive and interact with them. Platforms like X, Threads, Facebook, and Instagram have optimized their interfaces to connect people efficiently. I saw an opportunity to adapt these familiar interfaces for interacting with AI, allowing users to engage with multiple entities—whether human or artificial—in a seamless way. By leveraging the broadcast model user interface, I aimed to create a better method for people to engage with AI, enhancing the value they derive from these interactions. 2. Embracing Probabilistic Outputs for Richer Perspectives LLMs generate probabilistic outputs rather than singular, definitive answers. This fundamental characteristic challenges the traditional expectation of one objective response and invites a rethinking of how outputs should be delivered. SocialAI leverages this by providing multiple perspectives on a given question, acknowledging that a singular correct answer is often elusive or non-existent. By embracing the variability and randomness inherent in AI models, we offer users a range of insights that can sometimes provide unique value. This approach enriches the user experience, allowing for a deeper exploration of ideas and fostering a more nuanced understanding of complex topics. 3. The Value of a Private Space to Challenge Ideas Creating a private, non-judgmental space is crucial for users to comfortably explore and challenge their views. Unlike traditional social media platforms that often become echo chambers, SocialAI is designed to break that cycle by not merely validating ideas but actively challenging them. The private nature of the platform encourages open exploration without fear of embarrassment or criticism. This environment empowers users to think critically and engage in self-reflection, leading to personal growth and more informed decision-making. 4. The Importance of Transparency in AI Interactions Transparency is key in providing meaningful interactions between users and AI. By clearly indicating that users are engaging with AI and not humans, we can reduce emotionally charged reactions and promote more thoughtful, considered engagement. This clarity helps users focus on learning and decision-making rather than reacting to emotional triggers. Emphasizing transparency not only builds trust but also enhances the effectiveness of AI as a tool for empowerment, especially when navigating decisions without clear-cut answers. I truly couldn’t be more grateful. ❤️ After years of wanting to build such a product, I was finally able to make it a reality! Tens of thousands of people signed up & helped uncover all of these learnings! 🥰 What learnings or changed perceptions have y’all had since trying out SocialAI?