ML engineer and AI researcher building end-to-end ML pipelines, fine-tuning LLMs, and deploying anomaly detection systems. Currently pursuing an M.Sc. in Computer Science at Yeditepe University with thesis research on meta-learning for automated research paper analysis.
class BatuhanEdguer:
role = "AI Researcher & ML Engineer"
location = "Istanbul, Turkey"
education = "M.Sc. CS @ Yeditepe University (GPA: 3.47/4.00)"
thesis = "Meta Modeling For AI Solutions Based on Existing AI Configurations"
research = ["Meta-Learning", "LLMs & NLP", "Deep Reinforcement Learning", "HPC for ML"]
languages = ["Turkish (Native)", "English (Fluent)"]Teaching Assistant — Yeditepe University (Feb 2023 – Present)
- Delivering lab sessions for Data Structures, Software Engineering, and AI to 100+ students/semester
- Built automated grading pipelines, mentored 20+ students on ML & RL research projects
- Co-developed curriculum integrating PyTorch, HuggingFace, and LLM-based workflows
Software Engineer & AI Engineer — New Senses Space Technologies (Feb 2022 – Aug 2022)
- Developed anomaly detection models for satellite telemetry using end-to-end ML pipelines
- Designed data ingestion, feature engineering, model training, and deployment infrastructure
| Project | Description | Stack |
|---|---|---|
| Meta-Learning Research System | M.Sc. thesis — extracts metadata from AI papers & recommends optimal model configs. QLoRA fine-tuning (Qwen3-8B), active learning, multi-LLM annotation pipeline | FastAPI, Celery, PostgreSQL, Qdrant, Redis, MLflow, Docker |
| Bitcoin Price Prediction | NLP-driven prediction combining sentiment analysis on news with financial indicators. Benchmarked 18 ML models + CNN/LSTM | TensorFlow, scikit-learn, XGBoost, NLTK |
| RL Board Game Agent | Deep Q-learning agent for a 7x7 board game with experience replay, configurable network, PvP/AI modes | Keras, PyQt5 |
| Parallel Fractal Renderer | Mandelbrot/Julia renderer comparing OpenMP (CPU), CUDA (GPU), MPI (distributed) | C/C++, Python, PyQt5 |
| cc-vox | TTS plugin for Claude Code — spoken summaries with multi-backend & smart GPU awareness | Python, Qwen3-TTS, Fish Speech, CUDA |
| Caelestia | Contributor to a Hyprland-based Linux desktop environment (2.3k+ stars) | TypeScript, Shell |
Detailed breakdown
| Category | Technologies |
|---|---|
| Machine Learning | PyTorch, TensorFlow, Keras, HuggingFace, scikit-learn, XGBoost, LightGBM, vLLM |
| Languages | Python, C/C++, Java, TypeScript, SQL |
| ML Infrastructure | Docker, MLflow, FastAPI, Celery, Git, Linux, LaTeX |
| Parallel Computing | CUDA, OpenMP, MPI |
| Databases | PostgreSQL, Redis, MongoDB, Qdrant |
