Skip to content
View KimmoZAG's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report KimmoZAG

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
KimmoZAG/README.md

Kimmo Zhang · 张睿诚

LLM systems & AI infrastructure · Computer Science M.S. student at HUST

Agent memory · Inference orchestration · KV cache infrastructure

Portfolio · Publications · Contact · Try the LMCache calculator ↗

Selected work

NeuroMem · Streaming memory benchmark · ICML 2026, first author

Built a benchmark that interleaves writes and retrievals to measure how external LLM memory behaves as new information arrives.

Paper · Run the benchmark

SAGE · Dataflow orchestration · ICML 2026, co-author

Designed the DataStream API for composing LLM workflows and integrated NeuroMem as a backend for memory services.

Paper · Repository · Memory integration PR

LMCache · KV cache observability and capacity planning · Tencent internship

Worked on KV cache observability, trace replay, and capacity planning. Added Hunyuan and DeepSeek model support and fixed calculations in the open-source KV cache calculator.

Repository · Calculator PR

vLLM-HUST · LLM inference on Ascend

Worked on model loading and deployment tooling for LLM inference on Ascend hardware.

Repository

CANDOR-Bench · Streaming vector search · SIGMOD 2026

Contributed to research on continuous approximate nearest-neighbor search over evolving vector streams.

Paper · Benchmark code

Pinned Loading

  1. Intro4kimmo Intro4kimmo Public

    https://kimmozag.github.io/Intro4kimmo

    TypeScript

  2. LMCache/LMCache LMCache/LMCache Public

    LMCache: Supercharge Your LLM with the Fastest KV Cache Layer

    Python 12.4k 2.2k

  3. RIDE-Lab/SAGE RIDE-Lab/SAGE Public

    Streaming-Augmented Generative Execution, the IntelliStream ecosystem shared flagship product.

    Python 56 7

  4. CGCL-codes/neuromem-bench CGCL-codes/neuromem-bench Public

    Python