Highly analytical Data Scientist with a robust foundation in Mathematics and Statistics, backed by over 10 years of experience in education. Holds an M.Sc. in Statistics, with a thesis focused on predictive modeling and advanced analytics.
My professional journey is defined by a passion for transforming complex, unstructured data into actionable business intelligence through automation and machine learning.
- Languages & Core:
Python(Pandas, NumPy, SciPy),SQL,R - AI & NLP:
LLMs,RAG Pipelines,Agent,Langchain,NLP (NER),Model Evaluation - Data Engineering:
Web Scraping(Scrapy, Selenium, BeautifulSoup),Automated Pipelines,REST APIs,Supabase,MongoDB,PostgreSQL - BI & Visualization:
Power BI,Tableau,Looker Studio,Matplotlib,Seaborn - Cloud & DevOps:
AWS (Lambda, S3, EC2),GCP,Dokploy,Docker,FastAPI,Streamlit,Gradio
- Data Science: Specialized in building agentic workflows (Tagger, Extractor, Consolidator) and fine-tuning models for niche sectors like energy and clinical data.
- Automation: Expert in developing end-to-end automation systems, including large-scale data extraction and automated content management via WordPress REST APIs.
- Academic Focus: Completed M.Sc. thesis research on the comparative performance of 16+ machine learning models for advanced text classification.
- π M.Sc. in Statistics | Giresun University
- π B.Sc. in Mathematics Education | AtatΓΌrk University
- π« 10+ Years of Experience in Mathematics & Pedagogical Leadership
M.Sc. Thesis: Modeling of Data Obtained by Web Scraping Using Machine Learning Methods and Performance Comparison
Benchmarked 16 machine learning models (Transformers, RNNs, and traditional ML) on a 20K-article web-scraped maritime news dataset for multi-class business-event classification. BART achieved the best F1 Score (0.7955) among all models.
-
Creator | AI PR Reviewer (Jul 2026 β Present)
- Designed and shipped a free, bring-your-own-key GitHub App that automatically reviews every pull request for bugs, security issues, and performance problems β running entirely on GitHub Actions and costing $0 to operate, since it calls OpenRouter's free LLM models with each user's own key instead of a metered backend. Built the full pipeline solo: GitHub App auth, a Cloudflare Workers setup dashboard with a live free-model picker, encrypted secret provisioning, and automatic model fallback for rate-limited requests. Install it β
-
Data Scientist | John Snow Labs (Dec 2025 β Present)
- Focused on clinical data curation and building automated pipelines to enhance data processing efficiency by 30% through advanced NLP techniques.
-
S.Data Scientist | EnkiAI (Feb 2024 β Present)
- Leading NLP projects in the energy sector, implementing RAG pipelines and fine-tuning LLMs with a 20% improvement in response relevance.
-
S.Data Scientist | Westaco (Feb 2026 β Apr 2026)
- Improved and optimized an existing AI chatbot system by analyzing architecture, identifying performance bottlenecks, and enhancing response accuracy.
-
Freelance Data Scientist (Mar 2023 β Present)
- Developed end-to-end data solutions, including customer segmentation and sentiment analysis for various clients.
-
Mathematics Teacher | Secondary School (Apr 2017 β Present)
- Leveraged analytical skills to optimize student learning paths, resulting in a 40% increase in academic success metrics.
πͺπ‘ Strong analytical mindset, attention to detail, and effective communication skills add value to any data-focused team. Let's analyze the data together! ππ













