Skip to content
View lokashrinav's full-sized avatar
50 followers · 53 following

Achievements

Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: Pull Shark

Achievements

Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: Pull Shark

Block or report lokashrinav

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
lokashrinav/README.md

Shrinav Loka

About Me

Hi, I'm Shrinav Loka, a Computer Science and Math student at the University of Maryland - College Park. I previously interned at Stripe working on ML Infrastructure for Privacy. I'm super passionate about ML Systems, focusing on LLM inference, speculative decoding, KV cache mechanics, and prefill/decode disaggregation. Let's have fun coding!

Pinned Loading

  1. disagg-inference disagg-inference Public

    Python

  2. genserve genserve Public

    Python

  3. moe-serve moe-serve Public

    Python

  4. vllm-cold-start vllm-cold-start Public

    Cuts ~30s off vLLM cold starts by caching GPU initialization across container boots

    Python

  5. gpu-snapshot gpu-snapshot Public

    Multi-GPU checkpoint/restore for gVisor — 2 files, verified on 2x H100 with NCCL

    Shell