Zhichao Feng

Hong Kong, China

Join Prog.AI to see contacts

Summary

🤩

Rockstar

🎓

Top School

Zhichao Feng is an engineering professional with five years of experience bridging mechanical engineering research and data-driven software work, currently based in Hong Kong. With an MSc in Mechanical Engineering from The University of Hong Kong and a BEng from the University of Leeds, he has hands-on lab experience in tactile perception and material friction measurement as well as practical assembly and operations roles. In open source he contributed backend and evaluation improvements to the RecBole recommendation library, optimizing AUC calculations and adding GAUC tests to strengthen model evaluation. Known as a cooperative, self-motivated team player with strong communication skills, he also brings the discipline and energy of a university-level athlete to collaborative engineering challenges. An interesting detail: his background spans both physical measurement systems and recommendation-system metrics, giving him a rare cross-disciplinary perspective on empirical evaluation.

5 years of coding experience

The University of Hong Kong (HKU)

BEng (Hons), Mechanical Engineering, BEng (Hons), Mechanical Engineering at University of Leeds

English, Chinese, Chinese

Github Skills (7)

pytorch10

eval10

performance-optimization10

python10

evaluation10

scikit-learn9

scikit9

Programming languages (2)

Jupyter NotebookPython

Github contributions (5)

RUCAIBox/RecBole

Aug 2020 - Jan 2022

A unified, comprehensive and efficient recommendation library

Role in this project:

Back-end Developer & Data Scientist

Contributions:10 reviews, 114 commits, 17 PRs in 1 year 5 months

Contributions summary:Zhichao's contributions primarily focused on optimizing the evaluation process for the AUC metric within the recommendation system library. They modified the code in `evaluator/metrics.py`, `evaluator/utils.py`, and `evaluator/loss_evaluator.py`, indicating efforts to improve the efficiency of AUC calculations and potentially address calculation errors. Additionally, they added GAUC metric and implemented tests, demonstrating expertise in evaluating and refining model performance within the context of recommendation systems. These changes suggest a focus on improving model evaluation and refining underlying metrics.

pytorchrecommenderdeep-learningcollaborative-filteringsequential-recommendation

linzihan-backforward/RecSysEvaluation

May 2021 - May 2021

Code and Results for the paper: A Revisiting Study of Appropriate Offline Evaluation for Top-𝑁 Recommendation Algorithms.

Contributions:10 commits, 7 PRs, 2 pushes in 13 days

recommendation-algorithmsevaluationrecommendationsvdrecall

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.

Request Free Trial