Suggestions
Chris Lattner
Computer Scientist and Software Developer
Chris Lattner is a prominent American computer scientist and the co-founder and CEO of Modular AI, a company focused on developing an artificial intelligence developer platform. Born in 1978, Lattner is best known for creating LLVM, the Clang compiler, and the Swift programming language, which has significantly influenced modern software development.
Education and Early Career
Lattner earned his Bachelor of Science degree in Computer Science from the University of Portland in 2000, followed by a Master's degree in 2002 and a PhD in 2005 from the University of Illinois. His doctoral thesis was centered around LLVM, which he developed as a framework for optimizing compiler technologies.
Professional Journey
After completing his education, Lattner joined Apple in 2005, where he played a crucial role in developing various developer tools, including Swift, which was released in 2014. He later worked at Tesla as Vice President of Autopilot Software and held senior positions at Google, where he contributed to TensorFlow infrastructure and co-founded the MLIR compiler infrastructure.
In 2022, Lattner co-founded Modular AI with Tim Davis, aiming to simplify AI development and deployment through innovative infrastructure solutions. The company focuses on creating tools that make AI accessible to a broader audience by addressing the technical complexities that often hinder developers.
Contributions to Technology
Lattner's work has had a profound impact on both compiler technology and AI development. His initiatives have included:
- LLVM: A foundational compiler infrastructure that has become integral to many programming languages and systems.
- Swift: A programming language designed for ease of use while providing powerful capabilities for developers.
- MLIR: A compiler infrastructure aimed at reducing software fragmentation and improving compilation for diverse hardware.
Lattner's vision for Modular AI is to bridge gaps within the current AI ecosystem, making it easier for developers to create effective AI solutions without being bogged down by complexity.123
Highlights
Great to dive into PL with Richard Feldman on Software Unscripted (https://lnkd.in/gPQsXwWz), a legit language implementer! We explored a wide range of topics, incl lessons learned Swift=>Mojo, how the structure of integer types + overflow is more complex than it might look, approaches to safety, acceleration of novel HW with Mojo etc.
We had so much fun that there's a chance we might get to speak again in the future, let us know if there are specific topics you'd like us to dive deeper into on Software Unscripted!
YouTube video: https://lnkd.in/gcmFWJSY
Ad-free video for Patreon members! https://lnkd.in/gDPKyFfF
Audio version: https://lnkd.in/gaVbqN2B
It was super fun talking with ThePrimeagen and TJ last week: we relived the social side of launching Swift at Apple, lessons learned building scale projects, what Mojo🔥 is and how (yes) progress is still being made in systems, and more.
It was a lot of fun and they kept me on track and the discussion quite accessible. Thanks both!
