Tony Lee

About Tony Lee

I am having a lot of fun doing Android development, demo. OpenGLES app dev and demo on Android Tablet. Most complex programming issues solved in my career:* Debug memory leak issue happen when system scale up when OO database grow from 1MB -> 100MB -> 1GB.* Code, Debug and validate complex Interprocessing Communication between 44 CPUs (SMP and non SMP).* Debug 250ms delay in code path that triggers system switch over after system running in a few weeks.* Debug one issue with software system fail after system has been running in customer site for 250 days. Latest 2009 fun stuffs include opengl-es integration, H.264 HD encoder/decoder integration. Ubuntu Linux, tiny 1.2GHz ARM CPU.I love computer and network technologies range fromAndroid, Win32, COM/OLE.NET, embedded database, to Linux, Linux kernel. I am very fast learner and enjoy working at different computer/software fields love to use all kind of language, tools to produce better faster software.In addition to technology, I also like what technology can help people in everyday life. I like to see customers use and enjoy the products I help develop. I enjoy taking major ownership responsible and successfully deliver software products/projects for the complete product life cycle from designed, code, tested and maintain.I care greatly about software product quality. I prefer to cover the software features with automated regression test cases when possible. I like to make sure that computers/software/systems works 24/7 so I don\'t have to.I have strong experiences in how to handle complex programming issues such as SMP, cluster system performance, memory leak. I have obsession about getting microseconds/nanoseconds timing info about the software components I wrote.Specialties: Latest: Android, HC, GB, SDK, NDK, OpenGLES, ARM SOC, Old stuff: System Integration, C/C++, Expect, Embedded Linux, Web2.0, Python, TCL, Expect, MPEG, MPTS, HA (High Availability), digital video distribution system.

Experience

1. Sr Platform System Engineer

   AMD

   Sep 2021 — Present

   Sole creator/architect/developer of AGOF - AMD GPU Observability Framework. It monitor GPU server\'s thermal, power, etc, stats. Extremely useful for debug multiple DC thermal issues. The framework also use ebpf to collect call traces, timing info from GPU driver, Libraries\' APIs, user space code - very helpful to debug complex SW stack interactions issues for threads, processes. Debug issues related to PCIe bus, Video BIOS, System integration, Power, Thermal issues related thousands of racks deployed in one MDC customer / Multiple DCs.Debug MI300A/MI300X UBBFW / SMC Issues. Build/Test/Debug OpenBMC embedded linux ARM (ast2600) software. Help Debug MI300X deployment thermal/power/SW/HW issues at MDC (Mega Data Center) customer with failure case in < 10 system in thousands of racks deployments for months, by: 1) review detail low level logs, 2) review OpenBMC source code to isolate issue to SB-TSI arm linux kernel driver and i2c interfaces code. 3) Came up with test cases to validate the lower level failure states in failures systems. 4) Try various linux, driver, GPU level tests to recover. 5) Work with internal/customer team to test/debug/validate recover solution. Wrote a trace system to analyze the how build process of OpenBMC code base for MI300X UBB - to help me understand the files/process/program exec involved in building OpenBMC code base. Wrote Full Automated Test system that can run on hundreds of internal systems with test scripts that can test 1) Host Systems, 2) Virtual Machines. 3) Docker images. 4) BMC for CPU status, BIOS, Thermal, P, V, I, T states. 5) Collect UBB stats via ssh, scp, RF deep into the UBB system. Successful help MDC customer reach MP status with thousands of racks / 50k+ MI300X deployments after 1.5 years of hard works. Worked trace/profile setup on rocm/linux kernel/drive to visualize timing in nano-seconds accuracy.

Education

• University of California, Davis

  MSCS, MSCS, BSEE

• University of California, Davis

  MS, EE

  1986 — 1991

Skills

• Linux
• Android Bsp, Sdk, Ndk
• Distributed System
• Lamp
• Python
• Web Ajax
• Sqlite
• Video Conferencing
• Distributed Systems
• Shell Scripting
• C++
• Integration
• Embedded Systems
• Android
• Debugging
• Software Development
• Xml
• Testing
• Telecommunications
• Embedded Linux