I am a Machine Learning Engineer specializing in Deep Learning, Computer Vision, and Generative AI applications.
I build practical AI systems using Machine Learning, Deep Learning, Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG) architectures.
My interests include building intelligent applications, improving ML pipelines, and developing AI solutions that solve real-world problems.
- Building AI applications using LLMs, RAG architectures, and local AI inference
- Developing Farmer Helper AI, an offline agricultural assistant powered by Retrieval-Augmented Generation
- Improving machine learning systems through better evaluation, retrieval optimization, and deployment practices
- Exploring efficient AI deployment and production ML workflows
Machine Learning
- Supervised Learning
- Deep Learning
- Computer Vision
- CNN Architectures
- Transfer Learning
- Model Evaluation
- Large Language Models (LLMs)
- Retrieval-Augmented Generation (RAG)
- Vector Databases
- Semantic Search
- Hybrid Retrieval (BM25 + Dense Retrieval)
- Embeddings
- Cross Encoder Reranking
- Local LLM Deployment
- Data Processing
- Data Analysis
- Image Processing
- Visualization
- REST APIs
- FastAPI
- Software Architecture
- Git/GitHub
- Database Management
- Flutter Applications
- State Management
- Firebase Integration
- React Applications
- Responsive Web Development
- REST APIs
Cross-domain transfer learning for reliable condition monitoring of primary batteries under discharge-only operation
Measurement Science and Technology, 2026
Research on applying transfer learning techniques for reliable battery condition monitoring.
| Project | Description | Technologies |
|---|---|---|
| Farmer Helper AI | Offline agricultural AI assistant using RAG architecture. Provides farming guidance through document retrieval, embeddings, reranking, and local LLM inference. | Python, FastAPI, FAISS, BM25, Sentence Transformers, Ollama, Llama |
| Stroke Risk Prediction | Machine learning pipeline for predicting stroke risk using clinical data with preprocessing and model evaluation. | Python, Scikit-learn, Pandas |
| SynapseMNIST3D | 3D biomedical image classification using deep learning models. | Python, TensorFlow, Keras |
| ASL Recognition | American Sign Language recognition system supporting 29 classes with 98% accuracy. | Python, TensorFlow, CNN |
| Handwritten Recognition | CNN-based handwritten character recognition model achieving 97% test accuracy. | Python, TensorFlow, Keras |
| Study Group App | Secure cross-platform application for creating public and private study groups. | Flutter, Firebase |
| Food Delivery App | Mobile food ordering platform with dynamic menus and user interaction. | Flutter, Firebase |
| Turkey Travel Guide | Responsive tourism web application with database integration. | React.js, MySQL, REST API |
Explore more: GitHub Repositories