About me
Hello! Iām Danushka Liyanage, a data-driven problem solver with experience across industry and academia, and a Postdoctoral Research Fellow at the University of Sydney. My core interests lie in data-driven decision-making and the development of Agentic AI systems that can reason, adapt, and act reliably in complex environments.
My current postdoctoral research focuses on Agentic AI systems for healthcare, automated software testing, and benchmarking database systems, where I develop quantitative and statistical frameworks to support reliable decision-making under uncertainty.
Alongside my academic research, I bring extensive industry experience as a Data Scientist, delivering enterprise-grade solutions in predictive modelling, optimisation, and time-series forecasting across telecommunications, retail, and apparel domains. I have led and contributed to projects that translate complex data into actionable decisions, and I have hands-on experience building predictive and classification models using both classical statistical methods and modern machine-learning approaches.
I hold a BSc in Industrial Statistics (First Class Honours) from the University of Colombo, where I was awarded the Gold Medal for the best student in Industrial Statistics. I am now keen to apply my combined research and industry background to building decision-centric, agentic AI systems that deliver measurable real-world impact.
News
-
January 2026 — š„ Selected to serve as a program committee member for the short paper track of IEEE International Conference on Software Testing (ICST) 2026
-
January 2026 — š Our registered report titled "Evaluating Impact of Coverage Feedback on Estimators for Maximum Reachability in Fuzzing" has been accepted to Fuzzing 2026. Stay tuned for the pre-print.
-
January 2026 — š Invited to serve as a reviewer for Springer Nature - Automated Software Engineering
-
November 2025 — š Invited to serve as a reviewer for ACM Transactions on Software Engineering and Methodology (TOSEM)
-
September 2025 — š„ Invited to serve as a program committee member for FSE 2026
-
August 2025 — š Invited as an external reviewer to POPL 2025
-
A Benchmark for Databases with Varying Value Lengths
TPCTC - 2025
-
Assessing Reliability of Statistical Maximum Coverage Estimators in Fuzzing
ICSME - 2025
-
Extrapolating Coverage Rate in Greybox Fuzzing
ICSE - 2024
-
Reachable Coverage: Estimating Saturation in Fuzzing
ICSE - 2023 -
Estimating Residual Risk in Greybox Fuzzing
ESEC/FSE - 2021 -
Security Guarantees for Automated Software Testing
ESEC/FSE - 2021