Uploaded June 2026 | Updated September 2026, 3 weeks ago
Every model you use runs on tensors. But what actually is a tensor, where does DOOM fit in, and why does it matter when you are trying to get Copilot to optimize your code? Anthony Shaw breaks down how machine learning models work under the hood and what knowing that changes about how you write prompts that actually get results.
To learn more, please check out these resources:
* https://aka.ms/VSCode/Learn
* code.visualstudio.com/learn
๐ฆ๐ฝ๐ฒ๐ฎ๐ธ๐ฒ๐ฟ๐:
* Anthony Shaw
* Burke Holland
๐ฆ๐ฒ๐๐๐ถ๐ผ๐ป ๐๐ป๐ณ๐ผ๐ฟ๐บ๐ฎ๐๐ถ๐ผ๐ป:
This is one of many sessions from the Microsoft Build 2026 event. View even more sessions on-demand and learn about Microsoft Build at build.microsoft.com
LIVE165 | English (US)
Broadcast Stage
#MSBuild
Chapters:
0:00 - Exploring Model Example โ Harrier Text Embedding Model
00:03:59 - Discussion on Onyx being Turing complete and capable of running complex logic
00:04:31 - Reference to Doom running on Windows since 1995
00:06:26 - Audience poll concludes option 2 (CPU emulator) is correct
00:08:43 - Discussion on Excel interpreting machine code
00:10:37 - Explanation of Nectron representing a RISC CPU and RAM as tensor
00:13:56 - Scaling out challenges due to Doom being singleโthreaded
00:14:00 - Best practices for coaching AI agents โ importance of benchmarking
00:16:44 - Session conclusion and closing remarks praising the learning experience
Every model you use runs on tensors. But what actually is a tensor, where does DOOM fit in, and why does it matter when you are trying to get Copilot to optimize your code? Anthony Shaw breaks down how machine learning models work under the hood and what knowing that changes about how you write prompts that actually get results.
To learn more, please check out these resources:
* https://aka.ms/VSCode/Learn
* code.visualstudio.com/learn
๐ฆ๐ฝ๐ฒ๐ฎ๐ธ๐ฒ๐ฟ๐:
* Anthony Shaw
* Burke Holland
๐ฆ๐ฒ๐๐๐ถ๐ผ๐ป ๐๐ป๐ณ๐ผ๐ฟ๐บ๐ฎ๐๐ถ๐ผ๐ป:
This is one of many sessions from the Microsoft Build 2026 event. View even more sessions on-demand and learn about Microsoft Build at build.microsoft.com
LIVE165 | English (US)
Broadcast Stage
#MSBuild
Chapters:
0:00 - Exploring Model Example โ Harrier Text Embedding Model
00:03:59 - Discussion on Onyx being Turing complete and capable of running complex logic
00:04:31 - Reference to Doom running on Windows since 1995
00:06:26 - Audience poll concludes option 2 (CPU emulator) is correct
00:08:43 - Discussion on Excel interpreting machine code
00:10:37 - Explanation of Nectron representing a RISC CPU and RAM as tensor
00:13:56 - Scaling out challenges due to Doom being singleโthreaded
00:14:00 - Best practices for coaching AI agents โ importance of benchmarking
00:16:44 - Session conclusion and closing remarks praising the learning experience
