Akshay Agrawal
 |
Software Engineer, Google Brain Github /
Google Scholar /
Resume /
Blog |
Bio
I am a software engineer on the Google Brain team, where I work on TensorFlow, an open-source library for deep learning. My research interests include algorithms and domain-specific languages for mathematical optimization, machine learning (with an eye towards natural language processing), and distributed systems.
I graduated from Stanford University in 2017 with a Master's in theoretical computer science and artificial intelligence, a B.S. in computer science, and a minor in mathematics. During my Master's career, I joined the Information Systems Laboratory and conducted research in convex optimization under the supervision of Professor Stephen Boyd. I created a rewriting system for convex optimization problems, which we implemented in version 1.0 of CVXPY, a domain-specific language for convex optimization that is used by corporations and universities alike. A paper on our design is slated to appear in Journal of Control and Decision, 2018. During my undergraduate career, I collaborated with Andreas Paepcke to compile a tagged dataset of online educational discussion forum posts; we published our dataset and a machine learning intervention pipeline in the 2015 International Conference on Educational Data Mining. During both careers, I was fortunate to have Professor Mehran Sahami as a mentor.
If computer science is my first passion, then writing is my second: I served as a writer, news desk editor, and investigative news editor for The Stanford Daily, and I maintain a blog at debugmind.com.
Publications
A Rewriting System for Convex Optimization Problems A. Agrawal, R. Verschueren, S. Diamond, S. Boyd. To appear, Journal of Control and Decision, January 2018. 2017. (code)
YouEDU: Addressing Confusion in MOOC Discussion Forums by Recommending Instructional Video Clips. A. Agrawal, J. Venkatraman, S. Leonard, and A. Paepcke. Educational Data Mining. 2015. (code)
Technical Reports
A Cutting-Plane, Alternating Projections Algorithm for Conic Optimization Problems. A. Agrawal. 2017. (code)
Cosine Siamese Models for Stance Detection. A. Agrawal, D. Chin, K. Chen. 2017. (code)
Xavier : A Reinforcement-Learning Approach to TCP Congestion Control. A. Agrawal. 2016. (code)
B-CRAM: A Byzantine-Fault-Tolerant Challenge-Response Authentication Mechanism. A. Agrawal, R. Gasparyan, J. Shin. 2015. (code)
Teaching
I spent six quarters as a teaching assistant for the following Stanford courses:
EE 364a: Convex Optimization I. Professor Stephen Boyd. Spring 2016-17.
CS 221: Artificial Intelligence, Principles and Techniques. Professor Percy Liang. Autumn 2016-17.
CS 109: Probability for Computer Scientists. Professor Mehran Sahami and Lecturer Chris Piech. Winter 2015-16, Spring 2015-16, Winter 2016-17.
CS 106A: Programming Methodology. Section Leader. Lecturer Keith Schwarz. Winter 2013-14.
Industry
I work on TensorFlow on the Google Brain team. I honed my technical infrastructure skills over the course of four summer internships at Google, where I:
applied data science to conduct fleet-wide performance analyses of programs in shared servers and datacenters;
analyzed Dapper traces for the distributed storage stack and uncovered major performance bugs;
built a simulator for solid-state drives and investigated garbage reduction policies;
wrote test suites and tools for the Linux production kernel.
Writings
Optimization+ML Paper Summaries. Sumamries and commentary on papers that I find interesting. Feb. 2017-Present.
Graphical Model Sketch. Lecture notes on a sketch for graphical models. June 2016.
Max Stability and Misra Gries. Lecture notes on max stability and the misra gries algorithm. April 2016.
Learning about Learning: Machine Learning and MOOCs. A blog post about one of my applied machine learning projects. June 2015.
Machines that Learn: Making Distributed Storage Smarter. A blog post describing an initiative to make storage systems smarter via machine learning. Sept. 2014.