Jubi Taneja

Jubi Taneja is a Research Software Development Engineer at Microsoft, currently working on AI Compilers to generate compute-efficient kernels at the Research in Software Engineering (RiSE) group.13 She joined Microsoft in July 2021 and is based in the San Francisco Bay Area.1

Education and Background

Jubi holds a Ph.D. in Computer Science from the University of Utah, which she completed in 2020 with a GPA of 3.97/4.0.1 During her doctoral studies, she worked as a Research Assistant under the guidance of Prof. John Regehr, focusing on making compilers have more effective and reliable data flow analyses using formal semantics and solver-based techniques.1

Professional Experience

Before joining Microsoft, Jubi's career included:

• Research Intern at Mozilla, working on WebAssembly and superoptimization techniques for the Cranelift JIT compiler1
• Software Engineer and Senior Software Engineer at Samsung Research India - Delhi, where she worked on Linux Kernel Development1
• Research Fellow at IIT Bombay, implementing a generic dataflow analyzer in GCC1

Achievements and Recognition

Jubi has received several honors throughout her academic and professional career:

• Best Student Presentation Award at the ACM/IEEE International Symposium on Code Generation and Optimization (CGO) in 20201
• Best Paper Award at CGO 2020 for her work on "Testing Static Analyses for Precision and Soundness"1
• University Gold Medal from Punjabi University for securing the first rank in the Computer Science department1

Current Work and Interests

At Microsoft Research, Jubi is involved in:

• AI Compilers and generating compute-efficient kernels1
• Exploring the use of Large Language Models (LLMs) for loop vectorization2
• Superoptimizers and verification in the context of AI compilers3

Jubi is also active in the academic community, serving as a SIGPLAN Long-Term Mentor since June 2020, where she mentors Ph.D. students and software engineers across several countries.1