Han Li

Han Li is a Senior Machine Learning Engineer II specializing in Machine Learning Infrastructure at Instacart. Here are some key details about his professional background:

Current Role

Han Li currently works as a Staff Software Engineer at Meta (formerly Facebook) in the San Francisco Bay Area. Prior to this, he held the position of Software Engineer, Machine Learning Infrastructure at Instacart from June 2021.

Key Responsibilities at Instacart

As the tech lead of Instacart's ML Training Platform, Han Li:

• Led a multi-disciplinary team to design, develop, and launch Instacart's ML Training Platform
• Hosted production training jobs covering multiple ML frameworks (TensorFlow, PyTorch, XGBoost, LightGBM, CasualML) across all product lines
• Introduced Ray as the primary distributed computation framework for Instacart's ML Training Platform

Previous Experience

Facebook (September 2016 - June 2021)

Han Li worked as a Software Engineer on the Machine Learning Training Platform at Facebook for nearly 5 years. His key contributions included:

• Building PyTorch-based distributed training framework SDK and ML applications
• Designing the first generation of PyTorch model refreshing pipeline
• Leading platform consolidation for Caffe2-based online training on critical ranking products

Qualcomm (June 2015 - September 2016)

He worked as a Software Engineer, focusing on embedded LTE firmware algorithms for physical layer.

Education

Han Li holds a degree from The University of Texas at Austin, where he studied from 2013 to 2015.

Skills and Expertise

Based on his professional experience, Han Li has expertise in:

• Machine Learning infrastructure and platforms
• Distributed computing frameworks (e.g., Ray)
• PyTorch and other ML frameworks
• Software engineering for large-scale ML systems

Han Li's career trajectory shows a strong focus on machine learning infrastructure and distributed systems, with experience at major tech companies like Meta, Instacart, and Qualcomm.1