Haidong Lan
Machine Learning Engineer at MeshyAI
Shenzhen, Guangdong Province, China
Summary
🤩
Rockstar🎓
Top SchoolHaidong Lan is a Machine Learning Engineer with a decade of experience building high-performance inference systems for embedded, mobile and server environments. He led and shipped FeatherCNN at Tencent—an influential open-source CNN inference engine—optimizing it for aarch64, AVX servers and energy-efficient embedding systems used in gaming and AI medicine. Now at MeshyAI, he focuses on production-ready ML engineering that balances speed, resource efficiency and real-world integration. His background includes a PhD in Computer Science and research stints in bioinformatics sequence alignment, reflecting a strong blend of theoretical rigor and practical systems optimization. Notably, his contributions often target low-level build and model-loading robustness, showing attention to reliability as well as raw performance.
10 years of coding experience
3 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Shandong University
English, Chinese
Github Skills (7)
arm10
inference-engine10
c-language10
cprogramming-language10
neon10
performance-optimization9
machine-learning9
Programming languages (5)
Github contributions (5)
Tencent/FeatherCNN
May 2018 - Sep 2019
FeatherCNN is a high performance inference engine for convolutional neural networks.
Role in this project:
ML Engineer
ML EngineerContributions:1 release, 261 commits, 3 PRs in 1 year 4 months
Contributions summary:Haidong contributed to the FeatherCNN repository by fixing compilation errors, particularly for aarch64 Linux systems. They also addressed model loading issues, ensuring the successful loading of parameters and weights. Additionally, they modified and amended the booster files, suggesting an effort towards performance optimization and integration. The user's work indicates a focus on building and improving the inference engine's core functionality.
pytorchinference-enginedeep-learninginferenceconvolutional
taichi-dev/taichi_benchmark
Mar 2022 - Jan 2023
Contributions:5 reviews, 140 commits, 7 PRs in 9 months
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.