WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
World Congress 2026 Europe • Jul 10, 2026 • Session details

CUDA Python: GPU programming for the modern developer

Paul Graham

Achieve speed-of-light C++ performance without ever leaving Python. Learn how modern CUDA tooling unlocks massive multi-GPU acceleration and instant 10x speedups for your heaviest data workloads.

Pause
Mute Enter Fullscreen
#1 about 3 min

Introduction to CUDA and general-purpose GPU computing

A look back at the development of CUDA and the growth of hardware-accelerated scientific algorithms.

#2 about 2 min

Understanding modern data center GPU hardware architecture

How streaming multiprocessors and tensor cores process complex compute operations while managing thread latency.

#3 about 2 min

The expanded CUDA ecosystem and native Python support

Overcoming classical language barriers by integrating seamless programming compatibility across compilers and device libraries.

#4 about 2 min

Exploring the Python-enabled GPU programming software stack

How interoperable frameworks allow developers to smoothly transition between device environments and execution tools.

#5 about 4 min

Leveraging multi-node communication and core computation libraries

Utilizing specialized interfaces to orchestrate distributed cross-device computations without requiring CPU dispatch operations.

#6 about 2 min

Accelerating script execution with CuPy and Numba kernels

Replacing standard array packages with GPU-accelerated variants provides immediate speedups without extensive architectural knowledge.

#7 about 2 min

Framework-agnostic execution using the NV math library

A unified mathematics package automatically routes calculations to hardware components matching specific operational requirements.

#8 about 3 min

Optimizing GPU memory transfers through kernel fusion

Combining isolated data actions into a single operational instruction drastically reduces internal transfer overhead.

#9 about 2 min

Mitigating function dispatch overhead with CUDA graphs

Grouping short-running device instructions into structured execution sequences minimizes the latency of repetitive task launches.

#10 about 3 min

Performing high-level data transformations with cooperative libraries

Leveraging combined logical iterators and unified thread units improves computing performance when engineering chained reductions.

#11 about 5 min

Simplifying hardware abstractions using matrix tile programming

Delegating precise memory positioning to the compiler enables efficient hardware processing strategies when managing large data structures.

#12 about 2 min

Analyzing execution performance and debugging Python code natively

Identifying computational bottlenecks and programmatic flaws operates entirely within standard graphical integrated development environments.

#13 about 5 min

Navigating resource multiplexing and hardware compatibility questions

Understanding multi-instance scheduling behaviors prevents workload degradation and clarifies deployment constraints for production environments.

Matching moments

4:37 min

Simplifying parallel programming with the CUDA ecosystem

Paul Graham Paul Graham · LIVE

3:32 min

Evolution of general purpose GPU computing and Python

Paul Graham Paul Graham · World Congress 2025

2:27 min

Exploring the CUDA ecosystem and levels of abstraction

Paul Graham Paul Graham · World Congress 2025

2:18 min

Bridging the productivity and speed gap in Python

Andy Terrel · LIVE

1:37 min

Accelerating compute with focused developer tools

Julia Koch Julia Koch +1 · World Congress 2026 Europe

2:31 min

Defining CUDA as a comprehensive GPU platform

Stephen Jones · Coffee With Developers

More from World Congress 2026 Europe

Building the Nervous System of AI - Michael Kagan (NVIDIA) thumbnail
29:17
3 months ago

Building the Nervous System of AI - Michael Kagan (NVIDIA)

Michael Kagan, Sead Ahmetović

Five Things in Tech that Matter Now - WeAreDevelopers World Congress 2026 Closing Keynote thumbnail
19:41
3 months ago

Five Things in Tech that Matter Now - WeAreDevelopers World Congress 2026 Closing Keynote

Christian Heilmann

The Sound of Your Secrets: Teaching Your Model to Spy, So You Can Learn to Defend thumbnail
32:06
3 months ago

The Sound of Your Secrets: Teaching Your Model to Spy, So You Can Learn to Defend

David vonThenen

Super scaling for the Super Bowl: How to survive 30 million users hitting your backend in 30 minutes thumbnail
35:13
3 months ago

Super scaling for the Super Bowl: How to survive 30 million users hitting your backend in 30 minutes

Irina Branovic

Running Secure Life Science Research at Scale using Hybrid GPU HPC and Kubernetes 🧬 thumbnail
24:23
3 months ago

Running Secure Life Science Research at Scale using Hybrid GPU HPC and Kubernetes 🧬

Jeremy Murray

Angstfreude - AI and Corporate Culture - The Thrill and the Threat thumbnail
30:26
3 months ago

Angstfreude - AI and Corporate Culture - The Thrill and the Threat

Axel Ebert, Karin Krobath

Related videos

Accelerating Python on GPUs thumbnail
22:18
World Congress 2024 • about 2 years ago

Accelerating Python on GPUs

Paul Graham

Accelerating Python on GPUs thumbnail
24:39
World Congress 2025 • about 1 year ago

Accelerating Python on GPUs

Paul Graham

Accelerating Python on GPUs thumbnail
59:43
LIVE • over 2 years ago

Accelerating Python on GPUs

Paul Graham

CUDA in Python thumbnail
21:02
LIVE • over 1 year ago

CUDA in Python

Andy Terrel

Coffee with Developers - Stephen Jones - NVIDIA thumbnail
37:32
Coffee With Developers • over 1 year ago

Coffee with Developers - Stephen Jones - NVIDIA

Stephen Jones

WWC24 - Ankit Patel - Unlocking the Future Breakthrough Application Performance and Capabilities with NVIDIA thumbnail
22:07
World Congress 2024 • about 2 years ago

WWC24 - Ankit Patel - Unlocking the Future Breakthrough Application Performance and Capabilities with NVIDIA

Ankit Patel

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.