Skip to content
View Roozbehsn's full-sized avatar

Block or report Roozbehsn

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Roozbehsn/README.md

Hi, I'm Roozbeh Seyednozadi 👋

ML Engineer | LLMs & NLP Enthusiast


🚀 About Me

Hi, I'm Roozbeh Seyednozadi, a Computer Engineering student at IAUM (Islamic Azad University of Mashhad) in Mashhad, Iran.

I love coding, so I try to implement every concept I'm curious about, from LLMs and LLM architecture to NLP tasks like text cleaning and embedding algorithms such as BPE. I enjoy doing research in these areas too, which is why I've implemented two papers from scratch: it helps me build a deeper understanding while exploring the field further. I'm steadily working toward becoming an AI Engineer.

  • 🎓 Computer Engineering student at Azad University of Mashhad
  • 🧠 Focused on Deep Learning, NLP, and implementing research papers from scratch
  • 🤖 Deeply interested in the LLM world, how transformers, tokenization, and attention actually work under the hood
  • 🔭 I learn by doing: rebuilding classic ML/DL papers (Transformers, CNN ensembles, tokenizers) in pure PyTorch/Python instead of just reading about them
  • 🌱 Currently exploring optimization algorithms, NLP fundamentals, and going deeper into LLM-related projects
  • 📊 Also interested in core Machine Learning concepts, and I share notebooks on Kaggle as I work through them
  • 🔬 Research interests: Transformer architectures, attention mechanisms, tokenization methods (BPE), and NLP preprocessing techniques and other topics in this space
  • 📫 Reach me on LinkedIn
  • ✉️ Or email me at rseyednozadi@gmail.com

🛠️ Tech Stack


📖 Currently Reading

I enjoy reading books alongside coding, right now I'm working through:

Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow book cover

Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow


Thanks for stopping by! ⭐ Feel free to explore my repositories and connect on LinkedIn.

Pinned Loading

  1. Attention-Is-All-You-Need-From-Scratch Attention-Is-All-You-Need-From-Scratch Public

    Implementing the Transformer architecture from the “Attention Is All You Need” paper from scratch in PyTorch, including multi-head self-attention, positional encoding, encoder-decoder architecture,…

    Python 1

  2. MNIST-Ensemble-CNN-Paper-Implementation MNIST-Ensemble-CNN-Paper-Implementation Public

    PyTorch implementation of "An Ensemble of Simple Convolutional Neural Network Models for MNIST Digit Recognition" with M3, M5, M7 architectures, ensemble learning, and experimental results.

    Jupyter Notebook

  3. Text-Summarization Text-Summarization Public

    Extractive text summarization using two complementary approaches: a supervised BERT-based sentence classifier trained on CNN/DailyMail, and an unsupervised TF-IDF/TextRank pipeline applied to a cus…

    Python

  4. BPE-Tokenizer-from-Scratch BPE-Tokenizer-from-Scratch Public

    A Python implementation of the Byte Pair Encoding (BPE) algorithm from scratch for subword tokenization, demonstrated on a real-world text corpus using NLTK.

    Python

  5. NLTK-Text-Preprocessing NLTK-Text-Preprocessing Public

    A Python project demonstrating text preprocessing using the NLTK library, including custom regex tokenization, POS-aware lemmatization, and sentence segmentation for natural language processing tasks.

    Jupyter Notebook

  6. Particle-Swarm-Optimization-PSO-Visualizer Particle-Swarm-Optimization-PSO-Visualizer Public

    A Python implementation of the Particle Swarm Optimization (PSO) algorithm for continuous optimization problems. The project demonstrates swarm-based search using the Rastrigin and Shifted Griewank…

    Python