Zahra Moti

Zahra Moti

PhD Candidate
The Netherlands

I am a PhD candidate at Radboud University working on applying AI and large language models to analyze web and mobile ecosystems. My research combines large-scale web measurement, data collection, and AI techniques to detect privacy violations across websites and mobile applications. My research is published in leading academic venues such as ACM CCS and IEEE S&P.

Alongside my academic work, I collaborate with industry on applied privacy and AI systems. I currently work as a consultant with VaultJS (United States) and previously completed an industry internship at ING, where I built multi-agent retrieval-augmented pipelines for automated code understanding and modernization of large legacy codebases.

Education
2021 — Present

Ph.D. Candidate in Digital Security Group

Radboud University, The Netherlands
2017 — 2020

M.Sc. Computer Science

Shiraz University, Iran
2013 — 2017

B.Sc. Software Engineering

University of Isfahan, Iran
Work Experience
2021 — Present

PhD Candidate

Radboud University · Digital Security Group

Large-scale web measurement studies to detect online tracking and advertising using ML. Published at IEEE S&P and ACM CCS.

AI/MLLLMsData analysisWeb measurementsNetwork traffic analysisPythonJavaScriptGitHub
Jan 2026 — Present

Technical Consultant

VaultJS · United States (Remote)

Advising on AI-driven mobile app privacy analysis, including evaluation of LLM-based navigation agents for automated app exploration and security assessment.

Mobile privacyAILLMsMobileAWS
Apr 2025 — Oct 2025

Generative AI Engineer - Internship

ING · The Netherlands

Developed a multi-agent system with Retrieval-Augmented Generation (RAG) for automated code understanding and modernization.

LLMsRAGMulti-agentPythonSQLAzure
Oct 2023 — Dec 2023

Visiting Researcher

University of Twente · iSecLab

Developed an iOS automation library that leverages large language models and accessibility features to navigate mobile apps and analyze in-app advertisements.

PythoniOSLLMsAutomationPrompt EngineeringPrivacyText-to-speechVision Models
Sep 2020 — Sep 2021

Machine Learning Researcher

MarWell Bio · San Francisco (Remote)

Applied Deep Learning to therapeutic antibody discovery in a biotech research team.

Deep learningBioinformaticsPythonTensorFlowKerasGenerative Adversarial Networks
Publications
2026
LegacyTranslate: LLM-based Multi-Agent Method for Legacy Code Translation
Under review
2025
WhisperTest: A Voice-Control-Based Library for iOS UI Automation
ACM Conference on Computer and Communications Security (CCS)
2025

The Bitter Pill: Tracking and Remarketing on EU Pharmacy Websites

Data Privacy Management Workshop (DPM) — ESORICS
2025

Referrer Policy: Implementation and Circumvention

Proceedings on Privacy Enhancing Technologies (PETS)
2024

Targeted and Troublesome: Tracking and Advertising on Children's Websites

IEEE Symposium on Security and Privacy (S&P)
2022

VAEResTL: A Novel Generative Model for Designing CDR of Antibody for SARS-CoV-2

International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC)
2021

Generative Adversarial Network to Detect Unseen Internet of Things Malware

Ad Hoc Networks Journal
Skills

Machine Learning & AI

PyTorch TensorFlow HuggingFace Keras LLMs RAG GANs Sentence Transformers Prompt Engineering Agentic Systems

Privacy & Security

Web measurements Online tracking Advertising UI Automation Frida Network Traffic Analysis Malware detection

Engineering

Python JavaScript Java Puppeteer Pymobiledevice3 Selenium Playwright
Academic Service

Program Committee Member

  • The ACM Web Conference 2026
  • Workshop on Measurements, Attacks, and Defenses for the Web (MADWeb) 2026
  • European Workshop on Systems Security (EuroSec) 2026
  • USENIX Security (Poster session) 2024

Thesis Supervision

  • Leveraging LLM-based autonomous web agents for measuring personal data exfiltration in checkout forms
  • Security of local AI APIs and models shipped in major browsers such as Chrome
  • Agentic AI Browsers Under Threat: A Comparative Study of Vulnerabilities
Press Coverage
Contact