Cong Wang

Email: xiyou.wangcong@gmail.com

Summary

Innovative Software Engineering Manager with 16 years of industry experience in system technologies, including:

4 years leading a high-performance engineering team of up to 20 members
12 years of hands-on software development expertise
Driving innovation in Linux kernel, networking, and system performance optimization
Developing and implementing machine learning solutions for system technologies
Fostering open-source communities and aligning technical strategies with business objectives
Mentoring talent and cultivating collaborative engineering environments

Esteemed Linux kernel maintainer for the networking traffic control subsystem
Contributed over 1000 patches to Linux kernel
Proposed an innovative approach for integrating ML in Linux kernel: LPC 2023 Presentation
Pioneered eBPF Qdisc development: LPC 2020 Presentation
Invited speaker at industry-leading conferences:
- Netdev Conference 2021, 2023, 2024
- Linux Plumbers Conference 2020, 2023
- Open Source Summit NA 2024
- Netconf 2019
- LinuxCon Japan 2011, 2012
- LinuxCon Europe 2011

Leadership: Team Management, Project Management, Strategic Planning, Mentorship, Process Improvement
Technical: Linux Kernel Development, Networking (TCP/IP, RDMA), GPU, NCCL, eBPF, C, C++, Python
Domain Knowledge: System Technologies, Machine Learning for Systems, Performance Optimization

November 2020 - Present

Scaled the System Technologies and Engineering US team from 3 to 20 members
Conducted performance reviews and mentored team members, resulting in 5 internal promotions in 2 years
Led innovative Linux kernel projects, resulting in significant performance improvements:
- Directed eBPF sockmap development, contributing ~100 kernel patches and reducing CPU usage by 25% for metrics servers
- Pioneered eBPF Qdisc development, improving EDT bandwidth utilization from 74% to 99% under 5% packet loss
- Introduced ML for Linux kernel performance tuning, achieving 20% latency improvement for Nginx
Initiated system AIOps projects:
- Reduced chip macro placement iteration time from 2 weeks to 10 minutes using AI
- Optimized server racking with AI, improving power utilization by 30% and reducing human effort by 75%
Led a group of 5 engineers in developing and implementing high-performance Linux OS drivers, optimized libraries, and cutting-edge benchmarks for NVIDIA’s H100 GPU, enhancing system performance and compatibility.
Led a group of 3 engineers in optimizing internal C++ applications through advanced compiler tools and library enhancements, resulting in 1.3 million CPU core savings.

October 2013 - May 2020

Served as Linux kernel maintainer for the network traffic control subsystem
Contributed 500+ patches to the Linux kernel and various open-source projects
Led kernel development and performance optimization for Twitter OS team
Improved performance of Twitter Heron project by 4x and Twitter Memcache project by 2x
Designed and developed Mesos resource isolation for Twitter cloud platform

May 2009 - September 2013

Maintained kexec-tools and kdump for Red Hat and open-source community
Specialized in Linux networking subsystems: netconsole, bridge, bonding, TCP/IP stack, and openvswitch
Performed Linux kernel networking performance tuning for RHEL customers
Collaborated with security team to resolve Linux vulnerabilities
Presented at multiple Linux conferences, including LinuxCon Japan and Europe

June 2008 - December 2008

September 2004 - June 2008

Xi’an University of Post and Telecommunications, China