Sanjay Malakar

Computer Science Ph.D. Candidate at UC Riverside

I am a Ph.D. candidate in Computer Science at University of California, Riverside (UCR), advised by Prof. Manu Sridharan and a member of the RIPLE group. My research lies at the intersection of static program analysis and software reliability, with a focus on detecting, verifying, and repairing resource-management bugs in Java.

I am currently a Graduate Research Assistant at UCR. Previously, I worked as a Compiler Development Research Intern at Nissan North America, where I built compiler infrastructure for accelerator-based perception workloads. Before that, I worked as a Senior Software Engineer at OpenRefactory, Inc., developing SAST tools to detect and repair security and reliability bugs.

I hold an M.Sc. in Computer Science from UCR and a B.Sc. in Computer Science and Engineering from Bangladesh University of Engineering and Technology; my undergraduate thesis with Prof. Rifat Shahriyar focused on repairing data race conditions, and I also collaborated with Dr. Gias Uddin and Dr. Anindya Iqbal on empirical studies of low-code software development challenges.

I am particularly interested in tools that move from analysis to actionable repair.

Academic profiles: Google Scholar · ORCID · DBLP · UC Riverside

Recent News

Experience

Graduate Research Assistant

University of California, Riverside

Leading independent research on program analysis and automated repair of resource-management bugs. Developed Arodnap, an end-to-end repair tool for Java resource leaks in wrapper classes, increasing automated repair coverage from 41% to 68% on 285 open-source projects. Also developed a modular verification framework for resource leaks in collections of resources, demonstrating 62% precision and 100% recall on a 100-program benchmark.

July 2024 - Present

Compiler Development Research Intern

Nissan North America, Inc.

Worked with Nissan's Software-Defined Vehicles team on a compiler toolchain for accelerator-based perception workloads, building a Python -> Torch-MLIR -> MLIR -> LLVM/RISC-V pipeline. Designed a custom MLIR op with lowering to LLVM/RISC-V, including an LLVM intrinsic that maps to a custom RISC-V instruction. Built a runtime simulation workflow and demonstrated kernel-level throughput gains on representative GEMM workloads while maintaining correctness.

June 2025 - September 2025

Graduate Teaching Assistant

University of California, Riverside

CS 180: Introduction to Software Engineering

January 2025 - March 2025

Senior Software Engineer

OpenRefactory, Inc.

Contributed to the development of Intelligent Code Repair (iCR), a Static Application Security Testing (SAST) tool designed to detect and fix bugs in Java, Python, and Go projects. Key responsibilities and contributions:

  • Designed and implemented algorithms for pointer analysis, including incremental analysis and handling of threading and framework lifecycle methods.
  • Developed custom checkers for advanced bug detection, including taint analysis and null pointer detection.
  • Executed performance optimizations using ProtoBuf, improving memory efficiency and serialization performance.
  • Transitioned the deployment architecture from monolithic to microservices, enabling scalability for SaaS platforms.
  • Established CI/CD pipelines, integrating streamlined testing workflows and deployment automation.
February 2021 - August 2023

Education

University of California, Riverside

Doctor of Philosophy
Computer Science
September 2023 - Present

University of California, Riverside

Master of Science
Computer Science
September 2023 - June 2025

Bangladesh University of Engineering and Technology

Bachelor of Science
Computer Science and Engineering
February 2016 - February 2021

Publications

Verifying Resource Leak Freedom in Collections of Resources

Co-Authors: Sascha Kehrli, Michael D. Ernst, Martin Kellogg, Manu Sridharan

First modular static verification framework for resource leaks in Java collections of resources. Combines type and dataflow analysis to track disposal responsibility across collection updates and aliases.

Status: Under Review at ACM SIGPLAN OOPSLA 2026.

Repairing Leaks in Resource Wrappers

Co-Authors: Martin Kellogg, Michael D. Ernst, Manu Sridharan

End-to-end static analysis and repair pipeline for Java resource leak bugs in wrapper classes, improving automated repair coverage from 41% to 68% on 285 open-source projects.

Status: Accepted in 2025 IEEE/ACM Automated Software Engineering (ASE) Conference. [PDF] [CODE]

Developer Discussion Topics on the Adoption and Barriers of Low Code Software Development Platforms

Co-Authors: Md Abdullah Al Alamin, Gias Uddin, Sadia Afroz, Tameem Bin Haider, Anindya Iqbal

Published in the Empirical Software Engineering (EMSE) journal, this study builds on our MSR’21 paper. We applied topic modeling to 33K relevant Stack Overflow posts from 38 popular low-code platforms, conducting a manual annotation of statistically significant data samples to identify adoption barriers and trends.

Status: Accepted in 2023 Elsevier Empirical Software Engineering Journal (EMSE). [PDF] [CODE]

An Empirical Study of Developer Discussions on Low-Code Software Development Challenges

Co-Authors: Md Abdullah Al Alamin, Gias Uddin, Sadia Afroz, Tameem Bin Haider, Anindya Iqbal

This study analyzed approximately 5K Stack Overflow posts discussing nine popular low-code platforms. We identified trends, challenges, and implications for practitioners, platform providers, educators, and researchers.

Status: Accepted in 2021 IEEE/ACM 18th International Conference on Mining Software Repositories (MSR). [PDF] [Code] [Video]

RaceFixer - An Automated Data Race Fixer

Co-Authors: Tameem Bin Haider, Rifat Shahriyar

My undergraduate thesis, RaceFixer, is a Clang-based tool that automates the detection and repair of single-variable atomicity violations. It enhances ThreadSanitizer's bug reports with static analysis to generate suitable patches, emphasizing lock reuse and deadlock prevention.

Status: Published on arXiv. [PREPRINT]

Skills

Achievements

Scholarships

  • Deans Distinguished Fellowship - Awarded at the University of California, Riverside (2023-2024)
  • University Merit Scholarship - Achieved stipends from BUET in four out of seven terms for excellent result
  • Talentpool Scholarship - Achieved from Bangladesh Government in Higher Secondary School Certificate Examination (2016-2020), Secondary School Certificate Examination (2014-2015), and Junior School Certificate Examination (2011-2012) for outstanding performance

Awards

  • Dean's List Award - Got award from BUET in Junior year for an extraordinary result
  • Shapla Cub Award - Received from the honorable Prime Minister of Bangladesh for extraordinary performance as a Cub Scout

Contact

463 Winston Chung Hall, 900 University Ave. Riverside, CA 92521
iamsanjaymalakar@gmail.com
smala009@ucr.edu