I am a PhD student at Tel Aviv University, advised by Yonathan Efroni. I have broad interests in agentic AI and large language models. My current research focuses on studying reward hacking in LLM-based agents. Before that, I researched text-to-speech and speech language models at HUJI.

• HebTTS - A Language Modeling Approach to Diacritic-Free Hebrew TTS
• HebDB - A Weakly Supervised Dataset for Hebrew Speech Processing
• SALMon - An acoustic aware benchmark for evaluating speech language models
• SIMS - Scaling Analysis of Interleaved Speech-Text Language Models

• CTF - cybersecutiry low level ctf. buffer overflow, use after free and reverse engineering.
• ASR -Automatic Speech Recognition model developed as a final project in the course "Introduction to audio processing".
• VAE - Varational Auto Encoder, trained on MNIST.
• Velocity Selector - [physics] Analytical and numerical analysis of a velocity selector, written in python.
• Gravitational Lensing - [physics] Numerical analysis for gravitational lensing - causes from big masses wrapping the space around them.