https://turabiyldrm.com
Turabi Yıldırım
Software Engineer

About Me

I am a Computer Engineering graduate focused on Visual AI and remote sensing. My work spans 2D visual understanding, vision-language large models, and multimodal AI systems. I also enjoy exploring and testing different AI agent tools.

Recently, I have worked on satellite-image interpretation models, including both a frontier VLLM-based model and a parameter-efficient alternative.

Featured Projects

LLaVa-RS logo

LLaVa-RS

Frontier Vision-Language Model for Remote Sensing Image Understanding
MModalCC logo

MModalCC

Parameter Efficient Change Captioning Model
Coming Soon project logo

Coming Soon...

In Development

LLaVa-RS

2025 · VLLM · Remote Sensing

A remote-sensing VLLM project for interpreting satellite imagery and image-change captions.

  • Built a satellite-image interpretation VLLM based on LLaVa-OneVision.
  • Experimented with Qwen language models and SigLIP-style visual encoders.
  • Achieved top performance under current evaluation settings on a remote-sensing change-captioning dataset.
  • Presented the graduation project at SIU 2025 and published a related Hugging Face model.

MModalCC

2024 · PyTorch · JSTARS

Research work around multimodal satellite-image understanding, experiments, and publication support.

  • Trained and evaluated vision-to-text models for satellite-image understanding.
  • Tested visual augmentation methods including flipping and color transformations.
  • Worked on creation of SECOND-CC, a visual interpretation dataset for satellite-image understanding.
  • Contributed to research published in IEEE JSTARS, a Q1 journal with IF 5.3.

Coming Soon...

Coming Soon · AI Project

A new CUDA-focused project is being prepared and will be added here soon.

  • Project details are currently being finalized.
  • The work will be published after the first stable prototype.

Coming Soon...

Coming Soon · Robotics AI

A robotics-oriented AI project is planned for this slot.

  • Project details are currently being finalized.
  • The work will be published after the first stable prototype.

Current Focus

Avatar video generation logo Video Generation
Research Student
Yıldız Technical University · TÜBİTAK Research Group
Jun 2023 - Feb 2025

Developed PyTorch models from scratch, tested data augmentation techniques, and contributed to a JSTARS Q1 publication.

AI Intern
ASELSAN
Jul 2025 - Aug 2025

Worked on AI agent architectures and MCP; built a rapid unit-test generation prototype with LangChain, React, and WebSocket.

M.Sc. Computer Engineering
Istanbul Technical University Current GPA: 4.0
Feb 2026 - Present

Graduate study focused on computer engineering and AI-oriented research.

B.Sc. Computer Engineering
Yıldız Technical University GPA: 3.4/4.0
Sep 2021 - Aug 2025

Graduated with GPA 3.4/4. Relevant work included machine learning, algorithms, databases, networks, and systems.

Erasmus Exchange
University of Oviedo
Jan 2025 - May 2025

Studied C#, Java, Spring Boot, programming paradigms, OOP, and compiler design.

Contact

Let's build something great together.

Istanbul, Türkiye