I graduated from Sharif University of Technology in Computer Engineering in 2024, and I'm now pursuing my MSc in Computer Science (DSAI) at Saarland University. My current focus is LLM theory and AI safety, building on a background in Natural Language Processing, Machine Learning, and Human-Centered Computing.
Much of my published work sits in human-centered computing and NLP, from multimodal retrieval and accessibility to how people experience conversational systems. More recently, I've been moving toward understanding the foundations of large language models and how to make them safer and more reliable. Alongside research, I spent three years as a software engineer building products at scale.
Research Interests
Selected Research Projects
Personality Detection on Twitter (B.Sc. Project)
Detecting personality on X/Twitter using two Transformer modules for sentiment analysis over text and images, paired with a questionnaire-based pipeline to compare online signals against real-world personality.
Crosslingual Fact-Checked Claim Retrieval
Retrieving relevant fact-checked claims for multilingual social media posts. Proposed TriAligner, a dual-encoder pipeline that aligns native and English representations using contrastive learning, LLM-based augmentation, hard negative sampling, and GPT-4o reranking to improve monolingual and crosslingual fact-checking.
Multi-Agent Orchestration for Therapeutic Chatbots
An eight-day randomized controlled trial (N=66) comparing a multi-agent, finite-state chatbot with shared long-term memory against single-agent and unguided LLM baselines for self-administered psychotherapy. Showed that architectural orchestration, not prompt engineering alone, makes therapeutic dialogue significantly more natural and human-like.
Accessibility in Low-Code Applications via AI
Analyzing low-code application reviews to detect accessibility-related issues. Curated a novel review dataset with manual labeling and applied a Transformer-based classifier to separate accessibility concerns.
Publications
Experience
Industry
Teaching Assistant
Education
Awards
National University Entrance Exam (Konkour)
Ranked 47th among more than 150,000 participants in the Mathematics branch, Iran, 2019.
Skills & Languages
- FarsiNative
- EnglishFluent
- GermanBasic