The Modular community has been buzzing this month, from our Los Altos Meetup talks and fresh engineering docs to big wins with Inworld and Oracle. Catch the highlights, new tutorials, and open-source contributions in this edition of Modverse.

Blogs, Tutorials, and Videos

• Missed the Modular Meetup? Watch the talks online! Our Los Altos Modular Meetup brought together AI experts and builders for an evening of cutting-edge engineering insights. From democratizing AI compute to production-ready voice AI and matrix multiplication optimization, the talks are now available for everyone to watch. Talks include:‍
  
  • Chris Lattner, Modular – The future of democratizing AI compute and the role of open collaboration in accelerating progress.  
  • ​Feifan Fan, Inworld AI – How to integrate state-of-the-art voice AI into consumer applications and make it production ready, featuring insights from Inworld’s collaboration with Modular.  
  • ​Chris Hoge, Modular – Why matrix multiplication remains one of the most challenging problems in computer science and how the Modular stack helps developers optimize it.  

• At our September Community Meeting, we hosted a Q&A with the Mojo team to discuss the recently released Mojo vision and roadmap documents. We also heard from Bernardo Taveira on “Porting GSplat Kernels to Mojo” and Seif Lotfy on “Datastructures for DB Development.”

‍

‍

• ‍Modular x Inworld x Oracle. See how we helped Inworld slash TTS costs by 70% and boosted performance 4x by partnering them and Oracle Cloud—unlocking both NVIDIA and AMD GPUs for the first time. 

• Check out more internal engineering docs now available to the public: ‍
  
  • WGMMA Programming‍
  • Elementwise Operations on GPUs

• Explore our four-part series on matrix multiplication (matmul) kernels on NVIDIA Blackwell, where we walk through writing high-performance GPU kernels using Mojo, uncover advanced optimization techniques, and provide a step-by-step blueprint for maximizing NVIDIA’s Blackwell GPUs. ‍
  
  • Part 1: Introduction ‍
  • Part 2: Using Hardware Features to Optimize Matmul‍
  • Part 3: The Optimizations Behind 85% of SOTA Performance‍
  • Part 4: Breaking SOTA

Awesome MAX + Mojo

Highlights from the community’s latest projects:

• Slava created Stargine, a proof-of-concept game engine written in Mojo. ‍
• Yuhao Zhu built an online book on Mojo called “Mojo Miji - A Guide to Mojo Programming Language from A Pythonista’s Perspective”.
• Seif Lotfy implemented two variants of Bloom filters to dive deeper into Mojo. 
• Paul Dutton wrote LeNet-5* from scratch in Mojo without any additional libraries and trained on the MNIST data set.

Open-Source Contributions

If you’ve recently had your first PR merged, message the Modular team in the forum to claim your epic Modular swag! Check out the recently merged contributions from our amazing community members:

• ‍martinvuyk [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21]‍
• dl-alexandre [1]‍
• josiahls [1][2][3][4]‍
• rd4com [1][2]‍
• christoph-schlumpf [1] [2]‍
• gryznar [1][2]‍
• gabrieldemarmiesse [1]‍
• msaelices [1][2]‍
• mzaks [1]‍
• ThomasMader [1]

‍