Hamed Khademi Khaledi
AI & NLP Researcher · Tehran, Iran
Education
University of Tehran · Tehran, Iran
M.Sc. in Artificial Intelligence and Robotics
- GPA: 18.94 / 20.00
Shahid Beheshti University · Tehran, Iran
B.Sc. in Computer Engineering
- GPA: 18.78 / 20.00
Allameh Helli High School (SAMPAD) · Tehran, Iran
Diploma in Mathematics and Physics
- GPA: 19.62 / 20.00
Patents and Publications
- [C.1] Hamed Khademi Khaledi, Heshaam Faili (2024). IRUEX: A Study on Large Language Models Problem-Solving Skills in Iran's University Entrance Exam. Proceedings of the 31st International Conference on Computational Linguistics (COLING).
Academics
Graduate Research Assistant · University of Tehran
- Master’s thesis: Performance Evaluation and Enhancement of Large Language Models for the Iranian University Entrance Exam.
- Supervisors: Dr. Heshaam Faili, Dr. Mohammad Javad Dousti.
Undergraduate Research Assistant · Shahid Beheshti University
- Bachelor’s thesis: A Neural Network-Based Framework for Natural Language Processing in Persian.
- Supervisor: Dr. Hamed Malek.
Teaching Assistant · University of Tehran
- Course: Neural Networks — Instructor: Dr. Ahmad Kalhor.
Teaching Assistant · University of Tehran
- Courses: Neural Networks and Machine Learning — Instructor: Dr. Saeed Masoudnia.
Teaching Assistant · Shahid Beheshti University
- Course: Computational Artificial Intelligence — Instructor: Dr. Hamed Malek.
Teaching Assistant · Shahid Beheshti University
- Course: Discrete Mathematics — Instructor: Dr. Hamed Malek.
Teaching Assistant · Shahid Beheshti University
- Course: Advanced Programming — Instructor: Dr. Maedeh Mosharraf.
Experience
- Developed a specialized database for retrieving relevant texts efficiently.
- Analyzed different types of document retrieval methods.
- Developed an entity linking API for Persian Wikipedia.
- Trained a model to enhance entity linking accuracy in Persian.
Honors and Awards
- Irancell Labs artificial intelligence hackathon winner (Certificate) — Fall 2023.
- Ranked 3rd in bachelor's class (qualified to start a master’s degree directly, entrance exam requirement waived) — 2022.
- Ranked 1040th out of more than 140,000 attendees in Iran universities national entrance exam (Konkour) — 2018.
- Ranked 200th in Iran universities national entrance exam for master’s degree in artificial intelligence / computer science — 2022.
Skills
Programming Languages
Python, Java (OOP), MATLAB, C++, VHDL, Assembly X86, R
Data Science
Pandas, NumPy, TensorFlow, PyTorch, Jupyter
DevOps & Version Control
Git, Docker
Web Technologies
HTML, CSS, JavaScript, Angular, React
Research
Highly motivated in AI research; prepared for in-depth independent self-study
Other Tools & Technologies
Webots, Linux, LaTeX
Select Courses
During Master's
- Natural Language Processing (18.4 / 20)
- Foundation Models in Natural Language Processing (18.2 / 20)
- Deep Generative Models (18.3 / 20)
- Statistical Inference (19.5 / 20)
- Neural Network (19.69 / 20)
- Machine Learning (19.8 / 20)
- Data Analysis (19.04 / 20)
During Bachelor's
- Fundamentals of Robotics (19.92 / 20)
- Artificial Intelligence and Expert Systems (20 / 20)
- Principles of Algorithms (19.28 / 20)
- Software Engineering (15.75 / 20)
- Fundamental of Computational Intelligence (19.3 / 20)
- Engineering Probability & Statistics (18.5 / 20)
- Advanced Computer Programming (19.79 / 20)
Languages
- Persian: Native
- English: Professional work proficiency (TOEFL total: 100)
- Arabic: Elementary proficiency
Projects
HodHod Dataset · GitHub
Tools: Python, Multiprocessing
- HodHod is the first large-scale, cleaned Persian dataset in text format. Developed a comprehensive tool for end-to-end preprocessing of large Persian raw text datasets, streamlining tokenization, normalization, and data cleaning.
Piraye · GitHub
Tools: Python, Spacy, Nltk
- Piraye is a Python library that facilitates text normalization for Persian, Arabic, and English languages.
Persian NER & POS Plugin · GitHub
Tools: Python, PyTorch, HTML, CSS
- Developed part-of-speech tagging and named entity recognition models for the Persian language, achieving improvements over previous language models. Additionally, created a Chrome plugin to facilitate these tasks.
Iranian Music Dastgah Detection · GitHub
Tools: Python, PyTorch, Librosa, SVM, LSTM
- Developed a classification model to recognize the Dastgah of input music, applying advanced signal processing and machine learning techniques to identify Persian musical modes accurately.
Decaf Compiler · GitHub
Tools: Java, Lex
- Developed a compiler for Decaf, a simple object-oriented language similar to Java. The development process was structured into three phases: scanning, parsing, and code generation.
QA Bot · GitHub
Tools: Python, PyTorch, BERT
- Built a question-answering bot for the 2020 AI-Bot Challenge, implementing natural language processing techniques to understand and respond to user queries accurately.
Minesweeper · GitHub
Tools: MATLAB
- Developed a Minesweeper game using MATLAB, implementing game logic, user interface, and interactive features to recreate the classic puzzle experience.
References
- Dr. Heshaam Faili — Professor, University of Tehran · hfaili@ut.ac.ir
- Dr. Mohammad Javad Dousti — Assistant Professor, University of Tehran · mjdousti@ut.ac.ir
- Dr. Hamed Malek — Assistant Professor, Shahid Beheshti University · h_malek@sbu.ac.ir
Extracurricular
- Watching series and movies
- Listening to music
- Playing video games and board games
- Hiking, basketball, and soccer