Pengwei Song
Staff Data Engineer
Beijing, China
Summary
👤
Senior🎓
Top SchoolPengwei Song is a Staff Data Engineer based in Beijing with 11 years building large-scale data, storage and distributed systems for AI and graph workloads. He has delivered production infrastructure across Moonshot AI, NebulaGraph, MEGVII and Tencent—spanning pre-training pipelines, distributed transactions, consensus, PB-scale object stores and a MapReduce-like Python library for heterogeneous computing. Comfortable across storage engines, message queues and traversal/backup systems, he blends low-level systems design with data pipeline engineering to support AI at scale. His work often targets the intersection of performance and operability (e.g., execution plan pushdown, multi-step graph traversal and GC/index subsystems), and he shares technical insights on his Chinese blog.
11 years of coding experience
7 years of employment as a software developer
Master's degree, Computer Science and Technology, 15/221, Master's degree, Computer Science and Technology, 15/221 at 北京邮电大学
Chinese, Chinese
Github Skills (74)
parallel10
python10
graph-database10
hyperparameter-optimization10
mlops10
deep-learning10
automl10
computer-graphics10
ray10
cpp10
optimization10
leveldb10
llm-inference10
nebula-graph10
scalability10
Programming languages (6)
Github contributions (5)
songpengwei/blog-publish
Aug 2022 - Mar 2025
Contributions:104 pushes in 2 years 7 months
DistSysCorp/ddia
Feb 2022 - Jan 2023
DDIA 逐章精读
Contributions:5 reviews, 28 commits, 9 PRs in 11 months
ddiadistributed-systemsdatabasestorage
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.