Pouria Derakhshanfar

Postdoctoral Researcher

Delft University of Technology

My name is Pouria Derakhshanfar. I am a postdoctoral researcher in the Software Engineering Research Group (SERG) of the Delft University of Technology. I am currently involved in an H2020 funded project called DevOps for Complex Cyber-physical Systems (COSMOS). I am also part of the Computational Intelligence for Software Engineering Lab (CISELab). Earlier, I obtained my Ph.D. in Computer Science from the Delft University of Technology, where I defended my dissertation, entitled “Carving Information Sources to Drive Search-Based Crash Reproduction and Test Case Generation”. During my Ph.D., I was involved in another EU H2020 funded project, called Software Testing AMPlification (STAMP), where I developed new search-based approaches for crash replication and test amplification in a DevOps context.

My career goal is to combine human and artificial intelligence to implement tools and techniques to ease software development and testing tasks in industrial software projects.

Interests

Software Test Case Generation
Search-based Software Engineering
Software Engineering for Cyber-physical Systems
Software Test Amplification
AI for Software Engineering

Education

Ph.D. in Computer Science, 2021
Delft University of Technology
MSc in Software Engineering, 2017
Sharif University of Technology

Latest news

SSBSE 2021 paper is accepted (2021-06): Happy to announce that our research paper “Hybrid Multi-level Crossover for Unit Test Case Generation” got accepted at the 13th Symposium on Search-Based Software Engineering.
PhD defense (2021-04): I am honored to announce that I successfully defended my PhD. My thesis is available here.
COSMOS project kick-off (2021-01): We officially started the H2020 COSMOS project. In this project, we aim at addressing relevant challenges about the development and testing of complex Cyber-physical Systems.
SSBSE 2020 Best Paper Award: We are pleased to announce that we managed to win the best paper award for our paper It is not Only About Control Dependent Nodes: Basic Block Coverage for Search-Based Crash Reproduction.
ASE 2020 paper is accepted (2020-07): Super happy that our paper Good Things Come In Threes: Improving Search-based Crash Reproduction With Helper Objectives has been accepted at The 35th IEEE/ACM International Conference on Automated Software Engineering.
SSBSE 2020 challenge paper is accepted (2020-07): our challenge paper An Application of Model Seeding to Search-based Unit Test Generation for Gson got accepted at the 12th Symposium on Search-Based Software Engineering.
Two papers are accepted to SSBSE 2020 (2020-07): Our papers It is not Only About Control Dependent Nodes: Basic Block Coverage for Search-Based Crash Reproduction and Commonality-Driven Unit Test Generation have been accepted for publication in the proceedings of the 12th Symposium on Search-Based Software Engineering.
ASE 2020 tool demo accepted (2020-07): Our Botsing tool demonstration has been accepted at The 35th IEEE/ACM International Conference on Automated Software Engineering. ( demo video).
ICSE 2020 (2020-07): Our paper A benchmark-based evaluation of search-based crash reproduction will be presented in the journal first track of the 42nd International Conference on Software Engineering ( presentation video).
Crash fault localization master thesis (2020-06): S. Popping is defending his master thesis Automated crash fault localization.

Projects

DevOps for Complex Cyber-physical Systems (COSMOS)

An H2020 funded European project for designing and developing novel DevOps methodologies, techniques, and tools that will enable effective, continuous development and evolution of cyber-physical systems.

Follow Website

DevOps for Complex Cyber-physical Systems (COSMOS)

Software Tesing AMPlification (STAMP)

A collaborative effort between 10 organizations, from 6 European countries to advance the state of the art in DevOps and automatic software testing

Website Follow Follow

My Tools

Botsing (2018-current): An open-source search-based crash reproduction framework.
Cling (2019-current): An open-source search-based test generation tool for class integration testing.
ExRunner (2018-current): A bash-based infrastructure for running multiple instances of crash reproduction tools on 200 hard-to-reproduce crashes. ExRunner can be used to perform empirical evaluations of different crash reproduction techniques.