Pavle Subotic

Pavle Subotic PhD (UCL)

Intro

I am interested in researching and engineering solutions to large-scale industrial problems, in particular relating to Verifiers, Compilers, Databases, and Distributed Systems. I have a PhD from University College London (UCL)

I am investigating novel mechanisms to improve blockchain performance and security for the Sonic blockchain.
I was a Snr. Research Software Engineer at Microsoft. I primarily investigated and developed provenance and static analysis tools for data science.
I was a Snr. Applied Scientist at Amazon where I led the automated contract driven development process at Prime Video by founding and leading the development of the Coastguard tool among many other projects.
I was a visiting researcher at AWS where I founded and co-developed Tiros, a tool that analyses Ec2 networks for security vulnerabilities.
I spent a year at Oracle Labs where I worked on Souffle with my collegues Prof. Bernhard Scholz and Herbert Jordan. (See LPOP'22 talk here)
I was an R&D software engineer @ Optiver, Toshiba R&D and CSIRO. During my undergraduate studies I was a summer scholar at NICTA (now Data61) where I worked on hardware and processor simulation for the sel4 microkernal verification project.

Publications

Reusable Formal Verification of DAG-based Consensus Protocols, TBD'24
Nathalie Bertrand, Pranav Ghorpade, Sasha Rubin, Bernhard Scholz, Pavle Subotić (paper)

Abstract Interpretation-Based Data Leakage Static Analysis, TASE'24
Filip Drobnjakovic, Pavle Subotic, Caterina Urban (paper)

Ambit: Verification of Azure RBAC, CCSW@CCS'23
Matija Kupresanin and Pavle Subotic (paper)

Program Repair Guided by Datalog-Defined Static Analysis, ESEC/FSE'23
Yu Liu, Sergey Mechtaev, Pavle Subotic, and Abhik Roychoudhury (paper)

Bit-Vector Typestate Analysis, Formal Aspects of Computing
Alen Arslanagic, Pavle Subotic, Jorge Perez (paper)

Automatically Resolving Data Source Dependency Hell in Large Scale Data Science Projects, CAIN'23
Laurent Boué, Pratap Kunireddy, Pavle Subotic (paper) (ext paper)

Geyser: Provenance Extraction and Applications over Data Science Scripts, SIGMOD'23
F.Psallidas, M.Leszczynski, M.Namaki, A.Floratou, A.Agrawal, K.Karanasos, S.Krishnan, P.Subotic, M.Weimer, Y.Wu, Y.Zh (paper)

Efficient SMT-based Network Fault Tolerance Verification, FM'23
Yu Liu, Pavle Subotic, Emmanuel Letier, Sergey Mechtaev and Abhik Roychoudhury (paper)

Automatic Rollback Suggestions for Incremental Datalog Evaluation, PADL'23
David Zhao, Pavle Subotic, Mukund Raghothaman and Bernhard Scholz (paper)

Building a Join Optimizer for Souffle, LOPSTR'22
Samuel Arch, Xiaowen Hu, David Zhao,Pavle Subotic, Bernhard Scholz (paper)

Statically Detecting Data Leakages in Data Science Code, SOAP@PLDI'22
Pavle Subotic, Uros Bojanic, Milan Stojic (paper)

Scalable Typestate Analysis for Low-Latency Environments, iFM'22
Alen Arslanagic, Pavle Subotic, Jorge Perez (paper)

A Static Analysis Framework for Data Science Notebooks, ICSE'22
Pavle Subotic, Lazar Milikic, Milan Stojic (paper)

Towards Elastic Incrementalization for Datalog, PPDP'21
David Zhao, Pavle Subotic, Muhund Raghothaman, Bernhard Scholz (paper)

A Scalable Provenance Evaluation Strategy for Datalog, TOPLAS/POPL'21
David Zhao, Pavle Subotic, Bernhard Scholz (paper)

Specializing Parallel Data Structures for Datalog, CCPE Volume 34
Herbert Jordan, Pavle Subotic, David Zhao, Bernhard Scholz (paper)

Fast Parallel Equivalence Relations in a Datalog Compiler, PACT'19
Patrick Nappa, David Zhao, Pavle Subotic, Bernhard Scholz (paper)

Reachability Analysis for AWS-based Networks, CAV'19
Pavle Subotic, Byron Cook and many other authors (paper)

Brie: A Specialized Trie for Concurrent Datalog, PMAM@PPoPP'19
Herbert Jordan, Pavle Subotic, David Zhao, Bernhard Scholz (paper)

A Specialized B-Tree for Concurrent Datalog Evaluation, PPoPP'19
Herbert Jordan, Pavle Subotic, David Zhao, Bernhard Scholz (paper)

Automatic Index Selection for Large-Scale Datalog Computation, VLDB'19
Pavle Subotic, Herbert Jordan, Lijun Chang, Alan Fekete, Bernhard Scholz (paper)

Two concurrent data structures for efficient datalog query processing, PPOPP'18
Herbert Jordan, Bernhard Scholz, Pavle Subotic (paper)

Souffle: On Synthesising Program Analyzers, CAV'16
Herbert Jordan, Bernhard Scholz, Pavle Subotic (paper)

On fast large-scale program analysis in Datalog, CC'16
Bernhard Scholz, Herbert Jordan, Pavle Subotic, Till Westmann (paper)

Guiding Craig Interpolation with Domain-specific Abstractions, Acta Informatica, Issue 53
Jerome Leurox, Philipp Rümmer, Pavle Subotic (paper)

Horn Clauses for Communicating Timed Systems, HCVS'14
Hossein Hojjat, Philipp Rümmer, Pavle Subotic, Wang Yi (paper)

Exploring interpolants, FMCAD'13
Philipp Rümmer, Pavle Subotic (paper)

Logico-Numerical Max-Strategy Iteration, VMCAI'13
Peter Schrammel, Pavle Subotic (paper)

Technical Reports/Workshops/Abstracts

Beyond Deductive Datalog: How Datalog is used to perform AI tasks for program analysis, (Position Paper) LPOP'24,
Bernhard Scholz, Pavle Subotic and David Zhao (paper)

Towards Formal Verification of DAG-Based Blockchain Consensus Protocols (Extended Abstract) FMBC'24,
Nathalie Bertrand, Pranav Ghorpade, Sasha Rubin, Bernhard Scholz, Pavle Subotić (paper)

Designing a Datalog Engine for Industrial-Grade Static Analysis, (Position Paper) LPOP'23,
Pavle Subotic and Bernhard Scholz (paper, talk)

Rinser: Deriving Succinct Evidence for Automated Code Reviews, (Invited Talk) Infer Practitioners@PLDI'20,
Thanh-Toan Nguyen, Pavle Subotic, Bor-Yuh Evan Chang (paper, talk)

Incremental Datalog Prototype in Souffle, (Extended Abstract) IC@SPLASH'19,
David Zhao, Pavle Subotic, Bernhard Scholz
Static Analysis by Elimination, Bytecode@ETAPS'13
Pavle Subotic, Andrew Santosa, Bernhard Scholz (paper)

Enhancement of Teleconferencing Audio – Meta Information Extraction from Meeting Recordings. In Proc. CSIRO ICT Centre Conference'04,
Claudia Schremmer, Silvia Pfeiffer, Pavle Subotic, Eva Cheng