James Thomas

I am currently a sixth-year Computer Science PhD student at Stanford working with Professors Matei Zaharia and Pat Hanrahan on accelerating data-intensive applications. I received S.B. and M.Eng. degrees in EECS from MIT in June 2016.

CV, Scholar

Research Papers

1. My PhD research has been focused on productive but still performant hardware development flows for datacenter applications. Here is a paper on this work that appeared at ASPLOS 2020. A follow-up paper with Xilinx on speeding up compilation for large FPGAs is here.
2. Starting during my M.Eng., I worked on Weld, an intermediate language for data parallel computations that targets vector units, multicores, and GPU's. Here is a short paper on our work that appeared at CIDR 2017. Here is a longer paper in VLDB 2018.

Presentations

1. I have worked on a Pandas-based library for interpretable anomaly classification with GPU and FPGA acceleration: code, Data+AI Summit presentation (slides).
2. At Microsoft Research in the summer of 2015, I worked on a library for incremental computation on top of Apache Spark. My end-of-internship presentation can be found here.
3. At Cloudera in the summer of 2014, I worked on a number of improvements to the open-source Hadoop Distributed Filesystem (HDFS). A presentation I gave on my work can be found here.