I am a Software Engineer working as f Artificial Intelligence Engineer. My expertise lies in designing intelligent systems that are not only accurate but also efficient, scalable, and optimized for real-world deployment.
With a strong background in AI & Machine Learning (Generative AI, RAG pipelines), I specialize in bridging the gap between complex models and production environments. Unlike typical data science workflows, I focus heavily on system architecture, latency optimization, and robust engineering principles.
My technical foundation is rooted in C++ (C++17/20) and Embedded Systems. Having developed firmware for STM32 microcontrollers and real-time applications (FreeRTOS), I possess a deep understanding of memory management and bare-metal performance. This drives my strong interest in High Frequency Trading (HFT) and low-latency domains, where I aim to apply my skills in writing ultra-fast, lock-free code to solve complex algorithmic challenges.
Currently based in the Netherlands (search year visa) and open to roles in AI, Embedded Systems, or HFT where performance and engineering excellence are paramount.
These are some of my projects featured on this website. For more code and low-level implementations, check out my GitHub.
Embedded engineers waste hours searching through 2000+ page Reference Manuals. Finding specific Pinout or Register maps in complex tables is slow and kills the development flow.
I built a Mistral-powered RAG pipeline optimized for tabular data. It treats datasheets as a database, allowing engineers to query specs instantly with +90% faithfulness and +85% answer relevancy.
Text-based games used to be popular but disappeared due to lack of recognition of user input and getting into boring loops.
LLM integration for game management. The LLM increases creativity while keeping the main scenario of the game intact.
We want to reach enterprises to get clients for our digital consultancy. We need to find businesses with specific criteria (e.g., cafes in Arnhem without websites) to offer them services.
LLM-based searching with Google Maps data. This tool enables semantic search to find businesses matching specific criteria, making it easy to identify potential clients.
I wanted to find events in Antalya (my hometown) using semantic search instead of traditional keyword-based searches.
I scraped data from ticket vendor websites and used RAG (Retrieval-Augmented Generation) to enable natural language event discovery.
I build voice assistants for enterprises as a freelancer and needed a demo to send to potential customers.
Built this voice assistant with voice calling capabilities as a working demonstration of my voice AI solutions for businesses.
My non-technical background friends were having problems using the JobSpy repository.
I created this website to make the job search tool accessible to non-technical users with an intuitive interface.
Technologies and tools I work with, categorized by domain: