Conor Hoekstra - Parrot : Array GPU Programming

Why we picked it

**Abstract:** This talk introduces a new high-level CUDA C++ library that brings functional, array-style programming to GPUs - without compromising performance. With a fluent API that slashes boilerplate and minimizes kernel launches, it enables you to write expressive, high-performance GPU code with up to 10× less code. This talk explores how functional patterns can unlock both clarity and speed in GPU development. **Speaker: Conor Hoekstra** Research Scientist at NVIDIA working on array programming models and languages. Creator of the [Parrot](https://nvlabs.github.io/parrot/) library (a high-level, array-based library with implicit GPU fusion). Extremely passionate about programming languages, algorithms, and beautiful code. Founder and organizer of the [Programming Languages Virtual Meetup](https://www.meetup.com/Programming-Languages-Toronto-Meetup/). Host of four podcasts and an avid conference speaker.