Türkiye
Data Scientist with 4+ years of experience building production-grade machine learning systems in NLP, time series forecasting, and AI-driven products. Previously worked at Nesine, where I developed large-scale ML systems including content moderation models, multi-class classification systems, and time series forecasting models used for business-critical decisions. Experienced in designing end-to-end ML pipelines, from data collection and feature engineering to deployment and monitoring, often working with distributed systems such as Airflow and Kafka. Recently, I have been focusing on LLMs, RAG systems, and AI agents, building automation workflows and intelligent data-driven applications. Open to Data Scientist and AI Engineer opportunities.
• Built AI agents and LLM-powered automation systems to automate data workflows and decision-making processes • Developed end-to-end data pipelines combining web scraping, APIs, and ML models for real-time data collection and analysis • Implemented RAG-based systems using LangChain and transformer models for intelligent retrieval and context-aware responses • Designed ML-based systems for financial data analysis, including feature engineering, signal generation, and backtesting workflows
• Designed and built LLM-based QA systems using GPT models to improve user interaction and content understanding • Developed a LangChain-based system to generate dynamic, context-aware betting insights, enhancing user engagement and experience
• Built an NLP-based moderation system for the Kupondas platform to detect harmful user comments, achieving ~98% accuracy and significantly reducing manual moderation workload • Developed a large-scale multi-class classification system (1000+ categories) to automatically route customer requests, improving operational efficiency • Designed and deployed time series forecasting models to predict daily revenue, supporting financial planning and decision-making • Built liquidity forecasting models to estimate daily withdrawal demand across banks, optimizing capital allocation • Developed end-to-end ML pipelines including data ingestion, feature engineering, training, and deployment using Airflow, Kafka, and distributed data systems
• Designed and implemented a GAN-based synthetic image generation project using labeled urban datasets for a defense industry initiative • Organized and led a large-scale competition to generate realistic synthetic images, enabling the creation of high-quality training data • Contributed to a computer vision system for detecting counterfeit gold coins using image classification techniques
• Developed a real-time computer vision system to detect and analyze human head movements using deep learning models • Implemented image processing and deep learning techniques for real-time motion analysis via camera input
• Built NLP models using BERT for text classification, semantic understanding, and question answering • Developed systems capable of extracting meaning from text and generating accurate responses based on user queries