Keith Winstein is an assistant professor of computer science and, by courtesy, of electrical engineering at Stanford University. His research group creates new kinds of networked systems by rethinking abstractions around communication, compression, and computing. Some of his research has found broader use, including the Mosh tool, the Puffer video-streaming site, the Lepton compression tool, the Mahimahi network emulators, the gg lambda-computing framework, and the use of a temporal reordering threshold to detect packet loss. His work has received the Sloan Research Fellowship, the Usenix NSDI Community Award (2020, 2017), the Applied Networking Research Prize (2021, 2014), the Usenix ATC Best Paper Award, a Google Faculty Research Award (2017, 2015), a Facebook Faculty Award, the ACM SIGCOMM Doctoral Dissertation Award, and a Sprowls award for best doctoral thesis in computer science at MIT. Winstein previously served as a staff reporter at The Wall Street Journal, was one of the story consultants for HBO’s “Silicon Valley,” and worked at Ksplice, a startup company (now part of Oracle) where he was the vice president of product management and business development and also cleaned the bathroom. He did his undergraduate and graduate work at MIT.

Academic Appointments

2020-21 Courses

Stanford Advisees

All Publications

  • Learning in situ: a randomized experiment in video streaming Yan, F. Y., Ayers, H., Zhu, C., Fouladi, S., Hong, J., Zhang, K., Levis, P., Winstein, K., USENIX Assoc USENIX ASSOC. 2020: 495–511
  • Demo Abstract: RF Soil Moisture Sensing via Radar Backscatter Tags Josephson, C., Barnhart, B., Katti, S., Winstein, K., Chandra, R., IEEE IEEE. 2020: 365–66
  • Cracking open the DNN black-box: Video Analytics with DNNs across the Camera-Cloud Boundary Emmons, J., Fouladi, S., Ananthanarayanan, G., Venkataraman, S., Savarese, S., Winstein, K., ACM ASSOC COMPUTING MACHINERY. 2019: 27–32
  • From Laptop to Lambda: Outsourcing Everyday Jobs to Thousands of Transient Functional Containers Fouladi, S., Romero, F., Iter, D., Li, Q., Chatterjee, S., Kozyrakis, C., Zaharia, M., Winstein, K., USENIX Assoc USENIX ASSOC. 2019: 475–88
  • Multidecadal observations of the Antarctic ice sheet from restored analog radar records. Proceedings of the National Academy of Sciences of the United States of America Schroeder, D. M., Dowdeswell, J. A., Siegert, M. J., Bingham, R. G., Chu, W. n., MacKie, E. J., Siegfried, M. R., Vega, K. I., Emmons, J. R., Winstein, K. n. 2019


    Airborne radar sounding can measure conditions within and beneath polar ice sheets. In Antarctica, most digital radar-sounding data have been collected in the last 2 decades, limiting our ability to understand processes that govern longer-term ice-sheet behavior. Here, we demonstrate how analog radar data collected over 40 y ago in Antarctica can be combined with modern records to quantify multidecadal changes. Specifically, we digitize over 400,000 line kilometers of exploratory Antarctic radar data originally recorded on 35-mm optical film between 1971 and 1979. We leverage the increased geometric and radiometric resolution of our digitization process to show how these data can be used to identify and investigate hydrologic, geologic, and topographic features beneath and within the ice sheet. To highlight their scientific potential, we compare the digitized data with contemporary radar measurements to reveal that the remnant eastern ice shelf of Thwaites Glacier in West Antarctica had thinned between 10 and 33% between 1978 and 2009. We also release the collection of scanned radargrams in their entirety in a persistent public archive along with updated geolocation data for a subset of the data that reduces the mean positioning error from 5 to 2.5 km. Together, these data represent a unique and renewed extensive, multidecadal historical baseline, critical for observing and modeling ice-sheet change on societally relevant timescales.

    View details for DOI 10.1073/pnas.1821646116

    View details for PubMedID 31481619

  • Salsify: Low-Latency Network Video Through Tighter Integration Between a Video Codec and a Transport Protocol Fouladi, S., Emmons, J., Orbay, E., Wu, C., Wahby, R. S., Winstein, K., USENIX Assoc USENIX ASSOC. 2018: 267–82
  • Encoding, Fast and Slow: Low-Latency Video Processing Using Thousands of Tiny Threads Fouladi, S., Wahby, R. S., Shacklett, B., Balasubramaniam, K., Zeng, W., Bhalerao, R., Sivaraman, A., Porter, G., Winstein, K., USENIX Assoc USENIX ASSOC. 2017: 363–76
  • The Design, Implementation, and Deployment of a System to Transparently Compress Hundreds of Petabytes of Image Files For a File-Storage Service Horn, D., Elkabany, K., Lesniewski-Laas, C., Winstein, K., USENIX Assoc USENIX ASSOC. 2017: 1–15
  • The Case For Secure Delegation Kogan, D., Stern, H., Tolbert, A., Mazieres, D., Winstein, K., Assoc Comp Machinery ASSOC COMPUTING MACHINERY. 2017: 15–21
  • Network Stack as a Service in the Cloud Niu, Z., Xu, H., Han, D., Cheng, P., Xiong, Y., Chen, G., Winstein, K., Assoc Comp Machinery ASSOC COMPUTING MACHINERY. 2017: 65–71
  • Congestion-Control Throwdown Schapira, M., Winstein, K., Assoc Comp Machinery ASSOC COMPUTING MACHINERY. 2017: 122–28