Jiazhen Han
Senior Deep Learning Architect @ NVIDIA
United States
Greater Seattle Area
Computer Hardware
Software Development, C++, Linux, Python, CUDA, High Performance Computing (HPC)
Experience

Deep Learning Performance Architect Intern
United States
- Studied how a new type of prefetch mechanism can help DL inference, especially for the MLPerf-inference related networks. - Studied how the different optimizations enabled by CUDA Graphs can help the inference workloads, and what can we do to make it even faster. - Worked on a standalone Bert inference application, tried lots of crazy ideas that may bring better small batch performance on it.

GPU Compute Architect
Shanghai City, China
- Worked as a deep learning performance architect optimizing the performance of DL networks on NVIDIA's platform, including GPU servers and autonomous driving SoCs. - Constructed, calculated and prototyped new ideas that can bring better DL performance, mainly for DL inference operations that involve GPU grid-level concurrency (e.g. concurrent inferences, parallel branches, grouped operations, etc.). - Compiled the ideas, data, and prototypes into recipes for DL software teams (e.g. TensorRT, cuDNN, driver teams).
Jiazhen Han's Contact Information
Phone
Find the Right Leads
Find Verified Contact Data
What LeadContact does well
Find verified emails, phone numbers, and decision-makers with 98% accuracy.
Find Leads
Find the right people by company, role, industry, location, and more.
925M+ professional profiles

Find Emails
Access verified email addresses for your target contacts.
657M+ emails

Find Phone Numbers
Get cross-validated phone data from multiple top sources.
239M+ phone numbers

More Accurate. Lower Cost.
Find contact data in 1 tool with 98% accuracy
LeadContact integrates leading enrichment tools to deliver more accurate contact data—without paying for each one.
Great conversations start with the right contact.
It’s time to find yours.




