Last Update: Aug 2026
Download CV

vahyd@live.com

I am a First-year PhD Candidate at CySafe Lab, TU Delft, supervised by Dr. Savvas Zannettou and Dr. Yury Zhauniarovich.

My research focuses on safety in human–AI interaction, particularly Conversational AI Systems. I study how people interact with, perceive, and are affected by these systems, as well as the behavior of Conversational AI Systems themselves, with the goal of understanding and making these interactions safer. Grounded in a rigorous engineering background and professional experience as an AI/Software Engineer, I approach research with a builder's mindset.

My work has been published at ACL, EMNLP, NAACL, and AAAI ICWSM. I also regularly serve the research community as an Area Chair, Session Chair, and Reviewer for conferences including ACL, EMNLP, NeurIPS, AAAI, and ECIR.

On a personal note, I enjoy classic literature, especially Persian and Russian literature, and I was once a Dota 2 Immortal—a title I wear with mixed pride. Please don't hesitate to reach out if you see something interesting here :D I'm always open to new possibilities and experiences.

News

  • Aug 2026: Our paper Expectation, Backlash, Recovery, and Excitement: How Model Releases Shape Reddit Perceptions of Conversational AI Systems accepted to EMNLP 2026 Main Conference — explore the interactive paper companion
  • Apr 2026: Our paper Synthia: Scalable Grounded Persona Generation from Social Media Data accepted to ACL 2026 Main Conference
  • Nov 2025: Achieved IELTS Academic score of 8.5/9.0
  • Oct 2025: Started my PhD at TU Delft
  • Sep 2025: New resource paper, PolitiSky24 accepted to EMNLP 2025, Findings
  • Aug 2025: Defended my M.Sc thesis with distinction
  • May 2025: Our work on User Profiling in Social Networks with LLMs accepted to MisD@ICWSM 2025
  • Apr 2025: PerCul was presented in Frontiers 2025 symposium at TeIAS and won the honorable mention award

Education

Ph.D. in Artificial IntelligenceOct 2025 - Present

TPM @ Delft University of Technology

Working on the safe and responsible integration of AI in digital environments

Supervisors: Savvas Zannettou, Yury Zhauniarovich

M.Sc. in Information Technology EngineeringSep 2022 - Aug 2025

University of Tehran

GPA: 4.00/4.00 (18.5/20)

Thesis: User Behavior Modeling in Social Networks using LLMs

Supervisors: Azadeh Shakery, Masoud Asadpour

B.Sc. in Mechatronics, Automation & RoboticsSep 2016 - Sep 2021

Isfahan University of Technology

GPA: 3.11/4.00 (15.59/20)

Final Project: Android App for Planar Dynamic Balance of Rotors using Real-time Accelerometer Data

Publications

Expectation, Backlash, Recovery, and Excitement: How Model Releases Shape Reddit Perceptions of Conversational AI Systems

Accepted at EMNLP 2026 Main Conference · Interactive paper companion

V. Rahimzadeh, Y. Zhauniarovich, S. Zannettou

Synthia: Scalable Grounded Persona Generation from Social Media Data

Accepted at ACL 2026 Main Conference

V. Rahimzadeh*, E. Moosavi*, MT. Pilehvar, Y. Yaghoobzadeh *equal contribution

PolitiSky24: U.S. Politics Bluesky Dataset with User Stance Labels

Accepted at EMNLP 2025, Findings

P. Rostami, V. Rahimzadeh, A. Adibi, A. Shakery

PerCul: Story-Driven Cultural Evaluation of LLMs in Persian

Accepted at NAACL 2025 Main Conference

E. Moosavi*, V. Rahimzadeh*, Y. Yaghoobzadeh, A. Shakery, MT. Pilehvar *equal contribution

From Millions of Tweets to Actionable Insights: Leveraging LLMs for User Understanding

Accepted at MisD@ICWSM 2025

V. Rahimzadeh, A. Hamzehpour, A. Shakery, M. Asadpour

Notable Experiences

ResearcherOct 2024 - Oct 2025

Tehran Institute for Advanced Studies (TeIAS)

AI Engineer | LLM, LLMOpsOct 2023 - Oct 2024

RamonAI, ModAI App

  • Maintaining the main GitLab Repository with 7+ people actively working on it
  • Implementing & Monitoring LLM based Real-time pipelines on Millions of data using scalable queuing, databases & monitoring systems like NATs, Redis, PostgreSQL, Prometheus, Grafana & Sentry
  • Scalable Deployment & Monitoring of LLMs using vLLM, TGI, LangFuse, Grafana & Sentry
  • Using & Fine-tuning Different LLMs & VLMs
  • Implementing RAG based Solutions based on LLMs & KBs & Milvus & complex chatbot back-ends using LangGraph
  • Involvement in complex semantic/lexical search & retrieval systems using ElasticSearch

Co-Founder, Software EngineerJun 2019 - Oct 2023

Qorfechi

  • Planned, designed & developed multiple world class standard websites using the latest technologies
  • Implementing AI based features in Dashboards

Services

  • EMNLP 2026: Area Chair
  • ACL 2026: Session Chair
  • NeurIPS 2026: Reviewer
  • ECIR 2025-2026: Organizing Volunteer & PC Member
  • ACL 2025: Reviewer
  • AAAI 2025: Reviewer

Honors & Awards

  • Outstanding Paper Award at Frontiers 2025 symposium, Tehran
  • Ranked 2nd in Nationwide Graduate Level University Entrance Test of Information Technology
  • Ranked 6th in Nationwide Graduate Level University Entrance Test of Computer Engineering
  • Outstanding R&D Group of IUT in IoT & Robotics Group
  • Ranked among top 0.5% in Nationwide Bachelor Level University Entrance Test of Mathematics and Physics

Skills

Large Language ModelsVision Language ModelsResearchLangChainLangGraphMilvusElasticSearchNATsPostgreSQLRedisMongoDBvLLMTGIUnslothHF TrainerPyTorchBERTEmbeddersLexical RetrievalSemantic RetrievalColBERTBM25Neo4JCode ReviewingRepository MaintenanceTeamworkTight Deadlines

Interests

Fyodor DostoevskyAlbert CamusLeo TolstoyCormac McCarthyMilan KunderaMaxim GorkyFriedrich NietzscheCharles BukowskiFranz KafkaGabriel Garcia MarquezJordan PetersonAndrew HubermanLex FridmanSadegh Hedayat