Skip to content
View pkyria's full-sized avatar

Block or report pkyria

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.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
pkyria/README.md

I'm Panagiotis Kyriakidis πŸ‘‹

MSc Data Science | Machine Learning Engineer | Specialized in NLP

Based in Greece πŸ‡¬πŸ‡· | Speaks: English (C2), Greek (Native), German (B1)

LinkedIn Email

πŸš€ About Me

I am a Data Scientist currently pursuing my MSc in Data Science at the International Hellenic University. My background is in Computer Science, and I specialize in building scalable Machine Learning models, NLP systems, and data visualization pipelines.

  • πŸ”­ I’m currently working on AI & ML applications.
  • πŸŽ“ I previously researched Fake News Recognition for my BSc Thesis.
  • πŸ“Š I have experience in Digital Marketing Analytics and business intelligence.

πŸ› οΈ Tech Stack

Languages & Core: Python SQL Git

Machine Learning & AI: PyTorch TensorFlow Scikit-Learn Pandas NumPy RAG

Deployment & Tools: Docker FastAPI PySpark

Visualization & BI: Power Bi Tableau Matplotlib


πŸ“‚ Featured Projects

Project Tech Stack Description
MovieVec (Semantic Search & RAG System) Python, PyTorch, FAISS, FastAPI, Docker Built a semantic search engine for 800K+ movies using Sentence Transformers. Features vector databasing, cross-encoder re-ranking, and is deployed as a GPU-optimized service (<500ms latency).
Fake News Detection TensorFlow, NLTK, Scikit-learn Developed NLP models for fake news classification across three large-scale datasets. Achieved 78.33% accuracy via optimized text preprocessing and tokenization.
Customer Segmentation Python, Power BI, K-Means, RFM Team Lead. Developed K-Means and RFM segmentation models and churn forecasting. Created Power BI dashboards to improve targeting accuracy for marketing decisions.
Question Similarity Scikit-learn, PyTorch, S-BERT Built a duplicate question classifier using TF-IDF and Sentence-BERT embeddings. Compared XGBoost/LightGBM/LSTM models, achieving 0.26 log loss.

πŸŽ“ Education

MSc in Data Science | International Hellenic University (Thessaloniki) 2024 – Present

  • Focus: Machine Learning, Deep Learning, NLP, Big Data & Cloud Computing, Advanced Databases, Data Science for Business

BSc in Computer Science | University of Thessaly (Lamia) 2017 – 2023

  • Thesis: β€œSmart Recognition of Fake News”

Pinned Loading

  1. MovieVec-Semantic-Search MovieVec-Semantic-Search Public

    Python