Help & Supportcareers@gravityer.com
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Beijing University of Posts and Telecommunications
Data Scientist
June 25, 2026 – Present
Ok-Topk
November 27, 2021 – December 10, 2022
Ok-Topk is a scheme for distributed training with sparse gradients. Ok-Topk integrates a novel sparse allreduce algorithm (less than 6k communication volume which is asymptotically optimal) with the decentralized parallel Stochastic Gradient Descent (SGD) optimizer, and its convergence is proved theoretically and empirically.
View ProjectMagicube
October 25, 2021 – November 23, 2022
Magicube is a high-performance library for quantized sparse matrix operations (SpMM and SDDMM) of deep learning on Tensor Cores.
View ProjectDNN-cpp-proxies
September 10, 2021 – June 18, 2022
C++/MPI proxies for distributed training of deep neural networks.
View ProjectSpMV-on-Many-Core
July 2, 2021 – July 2, 2021
A cross-platform Sparse Matrix Vector Multiplication (SpMV) framework for many-core architectures (GPUs and Xeon Phi).
View ProjectChimera
May 30, 2021 – March 20, 2025
Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.
View ProjectFaiss_experiments
September 16, 2020 – July 29, 2025
Faiss_experiments — GitHub repository
View Projectdaceml
September 2, 2020 – December 14, 2025
A Data-Centric Compiler for Machine Learning
View Projectdeep-weather
May 8, 2020 – March 24, 2023
Deep Learning for Post-Processing Ensemble Weather Forecasts
View Projecteager-SGD
November 30, 2019 – November 18, 2021
Eager-SGD is a decentralized asynchronous SGD. It utilizes novel partial collectives operations to accumulate the gradients across all the processes.
View ProjectCultural Fit Analysis
The candidate's projects are heavily focused on deep learning infrastructure, distributed training, and performance optimization, which aligns well with a research-heavy Data Scientist role. However, the projects are all personal, and there is only one current work experience listed with a future start date, which limits the assessment of collaborative work in a professional setting. The breadth of application domains for data science is also limited to deep learning performance, rather than diverse analytical or modeling tasks.
Soft Skills & Operational Fit
Insufficient data to assess soft skills or operational fit. No psychometric or English test scores are available.