profile


Aritra Sengupta

Emails: aritra1686vit@gmail.com, sengupta.25@buckeyemail.osu.edu

  




Research Interests Publications
  Education
  About Me
News:
  • CompTaint, a paper on large-scale compositional static taint analysis to enforce policies on code, is accepted at ESEC/FSE 2023!
  • Presented RAPID at ESEC/FSE 2021, Aug 2021. Check out the talk to know how CodeGuru works at Amazon scale!
  • Our paper: RAPID on API checking in the cloud accepted at ESEC/FSE 2021!
  • An extended version of Transactuation accepted at ACM TOCS (2020)! 
  • Joined the Automated Reasoning Group at Amazon AWS!
  • Transactuation invited to be submitted to ACM Transactions on Computer Systems!
  • Transactuation wins the Best Paper Award at USENIX ATC 2019! Presented Talk at Seattle.
  • Transactuation accepted at Usenix ATC 2019 .
  • I joined Samsung Research America on July 2017.
  • Defended my PhD Dissertation: Efficient Compiler and Runtime Support for Strong Semantics on Commodity Hardware: Dissertation.
  • Paper on bounded region serializability using HTM accepted at CGO 2017. Presented the work at Austin on Feb 6, 2017.
  • Completed an internship at Microsoft Research. Worked with an amazing team on memory pressure aware server provisioning.
  • Presented a poster on bounded region serializability at OOPSLA 2015 student research competition.
  • Paper on enforcement of stronger memory model for Java accepted at PPPJ 2015. Presented the Talk at Melbourne, Florida.
  • Passed PhD candidacy examination.
  • Paper on enforcement of strong semantics using hybrid static-dynamic analysis accepted at ASPLOS 2015. Presented the paper at Istanbul, Turkey.
  • Presented a poster on a novel technique to enforce serializability at PLDI 2013 student research competition.
Research: 
       I am interested in work related to program analysis (static and dynamic analysis), parallel and distributed system software, and concurrency.

I am currently a Senior Applied Scientist at Amazon AI. I am working on checking/enforcing correctness properties on AWS services at scale. I have worked extensively on production-scale static analysis of code and JIT compilers. More recently, I have started looking into using PL to improve the quality of ML based code generation - enforcing responsible AI principles for code using PL and GenAI. I enjoy desigining and implementing analyses and systems considering tradeoffs between performance and correctness across the software stack.

Previously I was a Researcher at Samsung Research, in SRA's Artificial Intelligence Center, where I worked on building systems targeting reliability, performance, and explainability.

I earned my PhD. under the guidance of Prof. Michael D. Bond. Our research group at OSU: PLaSS. The doctoral research was directed towards low-overhead, state-of-the-art solutions that enforce strong semantics in ill-synchronized programs using hyrbid static-dynamic analysis, runtime support, compiler transformations, and software or hardware transactional memory, all on commodity hardware.

Resume: CV.
Refereed Conference and Journal Publications:

     Legend. AO: Authorship in alphabetical order of last names. If nothing is specified, the author names are in the order of contribution.

Patents:
 Technical Reports and Extended Versions:            
Service:

Education: I completed my Ph.D. at The Ohio State University focusing on programming language and runtime systems for reliable concurrent programs. I have done my Bachelor of Technology in Computer Science and Engineering from Vellore Institute of Technology, India.