Skip to content
View MuratKarslioglu's full-sized avatar
1 follower · 0 following

Block or report MuratKarslioglu

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
MuratKarslioglu/README.md

Murat Karslıoğlu

AI/ML Researcher — Quantum Computing × Large Language Models

Computer Engineering Student @ Sakarya University of Applied Sciences

Email Location

About

I work at the intersection of quantum computing and large language models — two fields I'm pursuing together toward a graduate degree in hybrid quantum-LLM research. Alongside this specialization, I have hands-on research experience across classical machine learning, reinforcement learning, metaheuristic optimization, and multi-criteria decision systems, giving me a broad base to draw on when designing hybrid or unconventional solutions.

current_focus   = ["Large Language Models", "Quantum-Classical Hybrid ML"]
also_working_on = ["Reinforcement Learning", "Metaheuristic Optimization", "Decision Support Systems", "Multimodal AI"]
goal            = "M.Sc. in Quantum Computing x LLM research"
  • 🔭 Currently interning on LLM research
  • ⚛️ Researcher at TÜBİTAK Milli Teknoloji Atölyesi — Artificial Intelligence & Data Science Center
  • 🧪 Research Assistant at the AI & Data Science Application and Research Center (Sep 2024–Present)
  • 🎯 Applying quantum-classical hybrid methods (VQE, QAOA, variational circuits) to optimization, RL, and hyperparameter search
  • 📊 ALES (Quantitative, 2025): 89.49
  • 📍 Istanbul, Türkiye

Core Expertise

⚛️ Quantum Computing

  • Variational Quantum Algorithms (VQE, QAOA)
  • Quantum Machine Learning (QML)
  • Quantum Reinforcement Learning (Quantum PPO, Quantum A3C)
  • Quantum circuit-based hyperparameter optimization — active research area
  • Quantum-classical hybrid architecture design
  • PennyLane, Qiskit
  • Real quantum hardware access: IBM Quantum, AWS Braket

🧠 Large Language Models

  • LLM research & applications (current internship focus)
  • Fine-tuning custom pipelines (Hugging Face Transformers)
  • Retrieval-Augmented Generation (RAG)
  • Local inference & experimentation via Ollama
  • Hybrid quantum-LLM research direction (graduate research goal)

Also Working On

Reinforcement Learning

  • PPO, A3C
  • Classical vs. quantum-hybrid RL benchmarking
  • Applied to packing, rocket landing, and control problems

Optimization & Algorithm Design

  • Genetic Algorithms
  • Chaotic systems (Hénon map, Lorenz system)
  • Hybrid classical/quantum optimization pipelines
  • Algorithm design, analysis, and performance optimization
  • Broad familiarity with optimization literature

Decision Science

  • Fuzzy Logic
  • TOPSIS & Multi-Criteria Decision Making (MCDM)
  • Decision-support system design

Multimodal AI

  • Vision-Language Models (VLM)
  • Speech-to-Text (STT)
  • Applied image and audio processing within LLM pipelines

Tech Stack

Languages

Python C C++ SQL

Quantum Computing

PennyLane Qiskit IBM Quantum AWS Braket

Deep Learning & LLM

PyTorch TensorFlow Hugging Face Ollama

Machine Learning & Data

scikit-learn NumPy Pandas SciPy XGBoost LightGBM

Data & APIs

JSON BSON REST API

Infrastructure & Tools

Docker AWS GCP Azure Vector DB Git Jupyter LaTeX

Featured Projects

3d-packing-quantum-classical-rl

3D container loading optimization comparing classical RL (PPO, A3C) against quantum-classical hybrid RL (Quantum PPO, Quantum A3C), using a VQE actor and QAOA critic circuit. Python PennyLane Reinforcement Learning

quantum-boosting-hyperopt

Quantum circuit-based hyperparameter optimization for boosting algorithms (XGBoost, AdaBoost, GradientBoost, CatBoost, LightGBM) using PennyLane variational circuits, benchmarked against classical tuning. Jupyter Notebook PennyLane Boosting

container-box-selection-chaos-topsis

Synthetic container-loading dataset generation using the Hénon map and Lorenz system, with TOPSIS-based multi-criteria selection. Python Chaos Theory MCDM

k2-exoplanet-detection

ML pipeline for exoplanet candidate detection from NASA K2 Mission stellar light curves — feature extraction, SMOTE balancing, Decision Tree / Random Forest classification. Jupyter Notebook scikit-learn

ElEm_RL_Rocket_Landing_Optimization

Reinforcement learning for rocket landing trajectory/control optimization. Python Reinforcement Learning

Contact

📧 karsliogluu00@gmail.com · 📍 Istanbul, Türkiye

Popular repositories Loading

  1. container-box-selection-chaos-topsis container-box-selection-chaos-topsis Public

    Synthetic container loading dataset generation using Hénon Map & Lorenz System, with TOPSIS-based selection

    Python 1

  2. 3d-packing-quantum-classical-rl 3d-packing-quantum-classical-rl Public

    3D container loading optimization using classical RL (PPO, A3C) and quantum-classical hybrid RL (Quantum PPO, Quantum A3C) with VQE actor and QAOA critic circuit

    Python 1

  3. MuratKarslioglu MuratKarslioglu Public

  4. quantum-boosting-hyperopt quantum-boosting-hyperopt Public

    Quantum circuit-based hyperparameter optimization for boosting algorithms (XGBoost, AdaBoost, GradientBoost, CatBoost, LightGBM) using PennyLane variational circuits — classical vs quantum comparison

    Jupyter Notebook

  5. k2-exoplanet-detection k2-exoplanet-detection Public

    ML pipeline for exoplanet candidate detection from NASA K2 Mission stellar light curves — feature extraction, SMOTE balancing, Decision Tree and Random Forest classification

    Jupyter Notebook

  6. ElEm_RL_Rocket_Landing_Optimization ElEm_RL_Rocket_Landing_Optimization Public

    Python