tduncan

tduncan

tduncan@lumieum.com

Activity Latest peak

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

Ascending recent streak, rightmost column is the freshest contribution.

Contribution Field

PO-inspired ramp using vertical u marks.

12 weeks

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

u

Older Recent high

about 1 hour ago

Created repository @tduncan/Cortana
about 1 hour ago

Created repository @tduncan/Nvidia-Optimization-Stack-VLLM
19 days ago

Pushed 3 commits to @tduncan/Nvidia-Optimization-Stack-VLLM on main
- 592c6f3fbc [Bugfix][Pooling] Handle multimodal cache misses without crashing (#58975)
- 443dccd3c9 [Bugfix][Multimodal] Fix reference counting of the SHM processor cache (#59695)
- c91dccc080 [Bugfix][Quark] Pass grouped-routing arguments to OCP MX monolithic kernels (#59752)
21 days ago

Pushed 3 commits to @tduncan/Cortana on main
- 0d43e44 refactor: support concurrent chat generation with multi-session management and UI indicators
- a74eeb5 feat: add ServerLogo component and integrate server brand logos across views
- de27783 feat(chat): keep a reply generating when switching chats #94