Mingji Han
Senior Machine Learning System Engineer at 腾讯
Boston, Massachusetts, United States
Summary
🤩
Rockstar🎓
Top SchoolMingji Han is a Senior Machine Learning System Engineer with a decade of experience designing and operating large-scale ML training platforms, currently building LLM training systems at Tencent after contributing to ByteDance’s Seed Foundation MLSys team. He combines strong distributed-systems and backend/DevOps skills—demonstrated by notable open-source contributions to CNCF project Thanos and the OpenTelemetry Collector, where he fixed data races, improved performance, and added robust storage and ingestion features. Mingji’s background spans production-grade checkpointing and streaming engines, including distributed checkpoint systems for LLM training and work on a next-generation streaming database. With an MS in Electrical and Computer Engineering from Boston University and a BE from Southern University of Science and Technology, he blends rigorous academic training with pragmatic engineering that surfaces in both internal platform work and respected open-source projects. An understated strength is his attention to low-level reliability (HTTP object-store tuning, multipart uploads, data-race fixes) that often prevents large-scale outages before they happen.
10 years of coding experience
2 years of employment as a software developer
Bachelor of Engineering - BE, Computer Science, Bachelor of Engineering - BE, Computer Science at 南方科技大学Master of Science - MS, Electrical and Computer Engineering, Master of Science - MS, Electrical and Computer Engineering at 美国波士顿大学
Github Skills (16)
opentelemetry10
kafka10
go10
prometheus10
devops10
clickhouse9
high-availability9
distributed-tracing9
storage9
cnc8
observability7
monitoring7
kubernetes6
docker6
kubernetes-pods6
Programming languages (8)
Github contributions (5)
thanos-io/thanos
Jul 2021 - Feb 2022
Highly available Prometheus setup with long term storage capabilities. A CNCF Incubating project.
Role in this project:
Back-end & DevOps Engineer
Back-end & DevOps EngineerContributions:38 reviews, 24 commits, 27 PRs in 7 months
Contributions summary:Mingji primarily contributed to the Thanos project by upgrading dependencies and adding features to the tools bucket web component, including time and relabel filtering. They also made significant changes to the COS object store client, adding HTTP configuration options and support for multi-part uploads to enhance performance and address potential issues with connection management. Moreover, the user addressed data races and improved performance within the query component, enhancing the stability and efficiency of the Thanos system.
promqlobservabilityprometheusstoragegoogle-cloud-storage
open-telemetry/opentelemetry-collector-contrib
Jun 2020 - Oct 2022
Contrib repository for the OpenTelemetry Collector
Role in this project:Backend Developer
Backend DeveloperContributions:95 reviews, 9 commits, 11 PRs in 2 years 4 months
Contributions summary:Mingji primarily contributed to the `kafkareceiver` and `jaegerreceiver` components within the OpenTelemetry Collector Contrib repository. Their work involved addressing data races, fixing typos, and adding support for strategy file auto-reloading in the Jaeger receiver. Additionally, the user added support for int counter metrics in the `deltatorateprocessor` and implemented log consumption logic within the `clickhouseexporter` module. These contributions indicate a focus on improving data ingestion, processing, and exporting functionalities.
contribopentelemetry-collectortracingtempoobservability
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.