As we wrap up August 2024, we’re excited to highlight some of the most insightful articles and posts shared by our vibrant community. Dive into these expert opinions, groundbreaking research, and practical guides to stay ahead in the AI and data science fields.

AI Trends in Data Science

Delve into the "AI Trends in Data Science" report, brought to you by NVIDIA and our community of 800 data scientists. This report covers the latest trends, key adoption barriers, and emerging opportunities. Enhance your AI strategy by exploring these unique insights. Check out the full report here.

Optimizing Language Models with NVIDIA

NVIDIA's research team has developed an innovative approach to creating smaller, highly accurate language models using structured weight pruning and knowledge distillation. This breakthrough reduces training time, costs, and improves model performance. Learn about how this technique was applied to the Llama-3.1-Minitron 4B model here.

Webinar: AI Studio Live Demo Recap

Our first webinar, "AI Studio Live Demo: Empowering Early Access Program Participants," was a huge success! If you missed it, don't worry—we've got you covered. Access the recording and additional resources to gain insights from the event:

• Webinar Recording
• AI Studio EAP Sign-up Form
• AI Studio Frequently Asked Questions

Most Popular AI Systems and Model Hubs Among Data Scientists & ITDMs

In collaboration with IPSOS, we've been looking to better understand the most commonly used Model Hubs and AI Systems among Data Scientists and IT Decision Makers (ITDMs).

Here’s a summary of our findings:

AI Systems:

• Data Scientists tend to use a wide range of compute systems, reflecting varied preferences across the group.
• On the other hand, ITDMs predominantly use Azure NDv4 and AWS EC2, with Google CSP and NVIDIA NIM also being commonly chosen.

Model Hubs:

• About half of Data Scientists use Microsoft AI GitHub to access pretrained models.
• Google Model Garden and AWS Model Zoo are also popular, ranking second and third, respectively.
• Interestingly, 23% of Data Scientists report not using any model hubs, indicating that AI adoption is still in its early stages in some organizations.

--

Thank you for being an integral part of our community. We’re looking forward to seeing more exciting contributions from all of you. As always, stay tuned for more updates, webinars, and insights coming your way!