1-day current streak·3-day longest streak
Machine Learning Systems, Alignment, and Evaluation I am a research-oriented machine learning systems engineer working on foundation model infrastructure, alignment, and evaluation. I build efficient and reliable systems for large…
Machine Learning Systems, Alignment, and Evaluation

I am a research-oriented machine learning systems engineer working on foundation model infrastructure, alignment, and evaluation. I build efficient and reliable systems for large language models while studying the algorithms and data choices that make them more useful, controllable, and cost-effective in real applications.
- 🧐 My central research interest is model-system co-design: understanding how model architecture, inference algorithms, data curation, hardware utilization, scheduling, and distributed runtimes interact.
- 💼 At TikTok, I work on Model-as-a-Service platforms and high-performance LLM inference, developing production serving infrastructure with vLLM and SGLang.
- 🎓 My recent research includes distributed disaggregated inference, preference optimization, instruction-tuning data selection, multimodal evaluation, and retrieval-augmented biomedical summarization.
- 🌱 I investigate alignment and evaluation methods that connect measurable model behavior with real-world usefulness, controllability, reliability, and serving cost.
- 📚 My systems work spans model runtime integration, scheduling and continuous batching, KV-cache and memory management, distributed execution, observability, and reliability.
- 💻 My broader research experience includes reinforcement learning for robotics, healthcare sequence modeling, privacy-preserving machine learning, and motion planning.
- ⛵ I am interested in collaborating on open research and infrastructure that make frontier AI systems faster to experiment with, more rigorous to evaluate, and dependable at scale.
- ✍🏻 I share technical writing on machine learning systems, infrastructure, and software engineering through my personal blog.
Some other achievements about me~e~e
- 💙💛 Be proud of the University of California, Berkeley. 🐻 Proud California Golden Bear. Fiat Lux ✨ Go Bears.
- 🌲 Be proud of Stanford University. ❤️ Proud Stanford Cardinal. Die Luft der Freiheit weht.
- 🧣 Be proud of Carnegie Mellon University. 🐾 Proud Carnegie Mellon Tartan. My heart is in the work.
- 🎉 Professional Membership of ACM / IEEE / IEEE-CS / CCF / Sigma Xi.
- 🍎 Apple Developer.👨🏻💻 & Apple Teacher.🤪
- 📊 Open-source activity and repository highlights:
Explore my repositories, research interests, and technical writing, or reach out to discuss machine learning systems and frontier AI.
<!---->
---
<!--
CSPs
Monitoring
Databases
-->
<!--
halfrost/halfrost is a ✨ _special_ ✨ repository because its README.md (this file) appears on your GitHub profile.
Here are some ideas to get you started:
- 🔭 I’m currently working on ...
- 🌱 I’m currently learning ...
- 👯 I’m looking to collaborate on ...
- 🤔 I’m looking for help with ...
- 💬 Ask me about ...
- 📫 How to reach me: ...
- 😄 Pronouns: ...
- ⚡ Fun fact: ...
-
Halfrost-Field ★ PINNED
✍🏻 这里是写博客的地方 —— Halfrost-Field 冰霜之地
Go ★ 13k 2y agoExplain → -
LeetCode-Go ★ PINNED
✅ Solutions to LeetCode by Go, 100% test coverage, runtime beats 100% / LeetCode 题解
Go ★ 34k 5h agoExplain → -
kubeedge ★ PINNED ⑂
Kubernetes Native Edge Computing Framework (project under CNCF)
★ 2 4y agoExplain → -
threes-ai ★ PINNED
🏆 Deep Reinforcement Learning for the Threes! game.
Go ★ 163 4y agoExplain → -
vue-objccn
🔥 Use Vue.js to develop a cross-platform full stack application / 用 Vue.js 开发的跨三端应用
Vue ★ 2.0k 4y agoExplain → -
Learn-Algorithms ⑂
算法学习笔记
C ★ 494 9y agoExplain → -
halfrost
🤫 You found a secret! halfrost/halfrost is a ✨special ✨ repository that you can use to add a README.md to your GitHub profile. Make sure it’s public and initialize it with a README to get started.
Go ★ 207 6d agoExplain → -
Prometheus
🌈 A Clean And Modern Ghost Theme with Progressive Web Apps (PWA)
CSS ★ 126 2y agoExplain → -
resume
🎖 my homepage (need VPN)
HTML ★ 47 7d agoExplain → -
Analyze ⑂
深入解析 iOS 开源项目
Objective-C ★ 47 8y agoExplain → -
The-Golang-Standard-Library-by-Example ⑂
Golang标准库。对于程序员而言,标准库与语言本身同样重要,它好比一个百宝箱,能为各种常见的任务提供完美的解决方案。以示例驱动的方式讲解Golang的标准库。
Go ★ 43 8y agoExplain → -
CS-APP3e
🌼 Homework of Computer Systems: A Programmer's Perspective (3rd Edition) and Autolab solutions of CMU 15-513: Intro to Computer Systems
C ★ 34 5y agoExplain → -
rust-learning
No description.
Rust ★ 30 5y agoExplain → -
halfrost.github.io
🚫 这里是冰霜曾经写博客的地方
Vim script ★ 24 5y agoExplain → -
weex-objccn
💎 这是一个 vue 转 weex 的项目
JavaScript ★ 22 9y agoExplain → -
S2
S2 geometry library in Go | experiment version
Go ★ 21 4y agoExplain → -
docker_practice ⑂
Learn and understand Docker technologies, with real DevOps practice!
Python ★ 19 8y agoExplain → -
2017_ele_hackathon_tank
🏆【2017 饿了么 黑客马拉松】 初赛第二题
Go ★ 17 1y agoExplain → -
BachGPT
No description.
Jupyter Notebook ★ 15 2y agoExplain → -
VueLearning
No description.
Vue ★ 11 5y agoExplain → -
awesome-rl ⑂
Reinforcement learning resources curated
★ 9 8y agoExplain → -
sentinel-go-envoy-proxy-wasm
Sentinel Go Proxy-WASM extension
Go ★ 8 3y agoExplain → -
astrolabes
some private config
Emacs Lisp ★ 6 5y agoExplain → -
starter-book
No description.
HTML ★ 4 3y agoExplain → -
langchain ⑂
🦜🔗 Build context-aware reasoning applications
Jupyter Notebook ★ 2 1y agoExplain → -
karmada ⑂
Open, Multi-Cloud, Multi-Cluster Kubernetes Orchestration
Go ★ 2 2y agoExplain → -
articles ⑂
Articles for objccn.io. objc.io的完整、准确、优雅的中文翻译版本
Objective-C ★ 1 8y agoExplain → -
vllm ⑂
A high-throughput and memory-efficient inference and serving engine for LLMs
★ 0 4mo agoExplain → -
sglang ⑂
SGLang is a fast serving framework for large language models and vision language models.
★ 0 7mo agoExplain → -
kubeflow ⑂
Machine Learning Toolkit for Kubernetes
★ 0 2y agoExplain → -
sentinel-go-ebpf-plugin ⑂
Sentinel eBPF plugin implemented by golang language
Go ★ 0 3y agoExplain → -
sentinel-go-proxy-wasm-extension ⑂
Envoy WASM extension with Sentinel Go
Go ★ 0 3y agoExplain → -
Pseudo-Delta-Neutral
Pseudo-Delta Neutral Strategy in Solana
Rust ★ 0 4y agoExplain → -
Poseidon
Haskell research
Haskell ★ 0 5y agoExplain →
No repos match these filters.