CV

This is a shortened version of my resume.

Basics

Name Sahar Admoni
Label PhD Candidate @ Technion – IIT
Email saharad@campus.technion.ac.il
Url https://saharadmoni.com
Summary My research focuses on the interaction between reinforcement learning and large language models, with the goal of improving sequential decision making and interpretability in AI systems. I study how LLMs can interpret and assist RL agents, and how RL techniques can support more consistent behavior and effective finetuning of LLMs.

Education

  • 2024 - Present

    Haifa, Israel

    PhD Candidate
    Technion - Israel Institute of Technology

    Thesis: “Leveraging Large Language Models to Interpret Reinforcement Learning Agents”
    Advisors: Prof. Ofra Amir and Dr. Assaf Hallak
    Key Collaborator: Prof. Yftah Ziser

  • 2023 - 2024

    Haifa, Israel

    MSc. as part of a direct PhD track
    Technion - Israel Institute of Technology
    The faculty of Data & Decision Sciences
  • 2019 - 2023

    Haifa, Israel

    BSc. Data Science & Engineering
    Technion - Israel Institute of Technology
    The faculty of Data & Decision Sciences

    Acquired theoretical foundations and hands-on experience in machine learning, deep learning, reinforcement learning, natural language processing, and statistical analysis.

Experience

  • 2025.09 - 2025.10

    Cambridge, MA, USA

    Visiting Researcher
    Harvard SEAS

    Collaboration with Prof. Finale Doshi-Velez at the Center for Human-Aware AI Research and Modeling.

  • 2024.06 - Present

    Israel

    Research Affiliate
    NVIDIA Research

    NVIDIA Research joined my PhD as a collaborative partner, providing mentorship, GPU resources, and co-supervision.

  • 2023.12 - 2024.07

    Daejeon, South Korea

    Research Scientist Intern
    KAIST & LG AI Research

    Developed a scalable methodology for evaluating subjective LLM metrics, contributing foundational research to the assessment pipeline for LG AI Research’s advanced language models.

Volunteer

Languages

English
Fluent
Hebrew
Native

Interests

Research
Reinforcement Learning (RL)
Large Language Models (LLMs)
Natural Language Processing (NLP)
Preference Optimization
Interpretability