Since November 2023 I am a postdoctoral researcher at TU Wien with Maria Christakis. I received my PhD from TU Munich, where I was co-advised by Eva Darulova and Helmut Seidl. Before that I was a graduate student in Max-Planck Institute for Software Systems (MPI-SWS) and Saarland University.

My research interests span over various topics in formal methods and software engineering. I am passionate about static analysis, in my book, AI stands for abstract interpretation. During my PhD, I focused on analysing and optimising numerical programs, now exploring new domains. I am curious about program synthesis, optimization, repair and testing and how they can help make software reliable and understandable.

News

• Our paper “Testing Static Taint Analyzers with Equivalence Modulo Taint” is accepted to ISSTA’26!
• I will give an invited talk about testing program analyzers at Helmutfest.
• I will be visiting Caterina Urban@Inria Paris and Raphaël Monat@Inria Lille in June 2026.
• I gave a talk about our work on the Cost of Soundness in Mixed-Precision Tuning at the FPTalks series on April 2.
• I am co-organising FLoC’26 Mentoring Workshop, looking forward to see you in Lisbon!

Service

Teaching

TU Wien

• Seminar in Software Engineering: SS24, WS24/25, SS25, WS 25/26
• Advanced Software Engineering WS25/26 - guest lecture

TU Munich

• Program Optimization: WS18/19, WS19/20, WS20/21, WS21/22, WS22/23
• Virtual Machines: SS19, SS20, SS21, SS23
• The Tyranny of Types: Curse or Blessing? (Seminar) SS22
• Program Synthesis Seminar: SS18, SS19, SS20
• Introduction to Informatics 2 (WS17/18)

Publications

• Testing Static Taint Analyzers with Equivalence Modulo Taint Maria Christakis, Anastasia Isychev, Samuel Pilz, Florian Tesarek, Valentin Wüstholz, ISSTA’26
• Interrogation Testing of CHC Solvers David Kaindlstorfer, Anastasia Isychev, Valentin Wüstholz, Maria Christakis, FSE’26
• Cost of Soundness in Mixed-Precision Tuning Anastasia Isychev, Debasmita Lohar, OOPSLA’25
• Using Action-Policy Testing in RL to Reduce the Number of Bugs Hasan Ferit Eniser, Songtuan Lin, Nicola Muller, Anastasia Isychev, Valentin Wustholz, Isabel Valera, Jorg Hoffmann, Maria Christakis, SoCS’25
• Lazy Testing of Machine-Learning Models Anastasia Isychev, Valentin Wüstholz, Maria Christakis, IJCAI’25
• Fuzzing Processing Pipelines for Zero-Knowledge Circuits Christoph Hochrainer, Anastasia Isychev, Valentin Wüstholz, Maria Christakis, CCS’25
• Interrogation Testing of Program Analyzers for Soundness and Precision Issues David Kaindlstorfer, Anastasia Isychev, Valentin Wüstholz, Maria Christakis, ASE’24
• Constraint-Based Test Oracles for Program Analyzers Markus Fleischmann, David Kaindlstorfer, Anastasia Isychev, Valentin Wüstholz, Maria Christakis, ASE’24
• Scaling up Roundoff Analysis of Functional Data Structure Programs Anastasia Isychev, Eva Darulova, SAS’23
• Regime Inference for Sound Floating-Point Optimizations Robert Rabe, Anastasiia Izycheva, Eva Darulova, EMSOFT’21
• Counterexample- and Simulation-Guided Floating-Point Loop Invariant Synthesis Anastasiia Izycheva, Eva Darulova, Helmut Seidl, SAS’20; Recording of the talk
• Synthesizing Efficient Low-Precision Kernels Anastasiia Izycheva, Eva Darulova, Helmut Seidl, ATVA’19
• Daisy - Framework for Analysis and Optimization of Numerical Programs (Tool Paper) Eva Darulova, Anastasiia Izycheva, Fariha Nasir, Fabian Ritter, Heiko Becker, Robert Bastian, TACAS’18
• On Sound Relative Error Bounds for Floating-Point Arithmetic Anastasiia Izycheva, Eva Darulova, FMCAD’17 (extended version on arXiv)

Theses

• Doctoral Thesis @ TU Munich Improving Analysis and Optimization of Finite-Precision Programs
• Master Thesis @ MPI-SWS and UdS Estimation of Relative Error Bounds for Floating-Point Arithmetic Expressions