- I am currently working at conway.ai building event-based foundation models
- I’m currently diving deeper into domain adaptive pre-training for pallas kernel specialization!
- Reach me @ sudarshan.jnambiar@gmail.com
founding eng @conway-engineering | comp eng @ purdue
- San Francisco
- @sdrshn_nmbr
Pinned Loading
-
inkle
inkle PublicThinking Machines Inkling inference on TPUs via JAX — checkpoint IO, NVFP4 decode, on and v7x's
Python
-
palinkle
palinkle PublicThinking Machines Inkling Small and Poolside Laguna XS2.1 post-training and specdec via SGLang and Miles, Pallas kernel bench (including megakernels for KDA, DeltaNet, etc.), specdec profiling for …
Python
-
auroraplusplus
auroraplusplus Publiccontinuously improving draft models for spec dec, mainly pulling inspiration from https://cursor.com/blog/real-time-rl-for-composer
Python
-
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.



