Nice to meet you,
I am a 3rd-year computer engineering student with a passion for artificial intelligence and machine learning. I was a finalist in the 2024 Teknofest Health AI competition, a milestone that underscores my commitment to developing innovative solutions in fields such as large language models and image processing.
I am Barış
EXPERIENCE
AI & LLM Backend Engineer (Remote, 2024 - Present)
Developing an agent-based AI system using Python, Docker, FastAPI, and Firebase.
Implementing SmallAgent and RAG frameworks for multi-language intelligent automation.
Optimizing real-time data processing and model deployment pipelines.
AI Engineer Intern – Dpointgroup (Barcelona, 2024)
Fine-tuned LLaMA and Qwen models for task-specific NLP adaptations.
Processed and optimized company data using Pandas, NumPy, and Variance Threshold.
Developed Word2Vec embeddings for text data representation and automation scripts.
Google Core Team Member | Kütahya (2023 - Present)
I am a Google Developer Groups (GDG) Kütahya Core Team Member, contributing to community events, workshops, and AI-focused initiatives.
SKILLS
Programming & AI: Python, TensorFlow, Scalable ML (Apache Spark), Docker
Computer Vision: Object Detection & Recognition (YOLO, Faster R-CNN), OpenCV, Image Processing
Machine Learning & NLP: Feature Engineering, Hadoop, Pandas, NumPy, RAG, Fine-tuning LLMs, Multi-Agent Systems, Large Language Models
Backend & Deployment: FastAPI, Firebase, Streamlit, Flask
CERTIFICATIONS
Agentic AI and AI Agents: A Primer for Leaders – Vanderbilt University (March 2025)
Milli Teknoloji Akademisi Yapay Zeka Temel Eğitim Sertifikası – T.C. Sanayi ve Teknoloji Bakanlığı (Feb 2025)
Introduction to Docker – Google (Jan 2025)
IBM Deep Learning & Neural Networks with Keras – IBM (Sept 2024)
Machine Learning with Python – IBM (May 2024)
Python OpenCV ile Sıfırdan Uzmanlığa Görüntü İşleme – Datai Team (Feb 2024)
Generative AI with Large Language Models – Coursera (Feb 2024)
C | Sıfırdan İleri Seviyeye Komple C Programlama – Udemy (Nov 2023)
GNU/Linux System Administration Level 1 – Linux Kullanıcıları Derneği
Teknofest 2024 Finalist – Health AI Competition, Disease Detection via Computer Vision
📄 Certificate Verification
🔗 Conference Presentation: Modular and User-Friendly Interface Development for AI-Assisted Animal Behavior Experiment
📄 11th International Congress on Life, Engineering, and Applied Sciences in a Changing World Proceedings Book
Freelance WordPress Developer (2019 - Present)
Designing and developing custom WordPress websites for clients across various industries.
Few of my projects
PalmReaderAI
Palm Reader AI is a web application developed using Streamlit, leveraging the Llama-3.2-90B Vision model for advanced palmistry analysis. The system processes high-resolution palm images, extracting key features such as line patterns, ridge structures, and shape classifications.
Feature Engineering for Machine Learning
The Feature Engineering for Machine Learning project focuses on optimizing datasets for improved model performance. It includes techniques such as variance thresholding, scaling, encoding categorical variables, and dimensionality reduction to enhance ML model efficiency. The project is implemented in Python using libraries like Pandas, NumPy, and Scikit-learn, ensuring scalable and structured data preprocessing.
Spark California Housing Regression
The Spark California Housing Regression project leverages Apache Spark MLlib for distributed regression modeling on the California Housing dataset. It applies feature engineering (handling missing values, normalization, and transformations) and trains Linear Regression and Random Forest Regressor models. The pipeline is optimized for parallel processing, ensuring scalability for large datasets. The final model achieved an R² score of 0.797 and an RMSE of 44,570, demonstrating robust predictive performance for housing price estimation
Teknofest 2024 Health AI Finalist
The Teknofest 2024 Health AI project focuses on breast cancer detection using YOLO-based lesion identification and BI-RADS classification. It integrates two LLaMA-based LLMs for radiology report correction and fine-tunes models for improved accuracy. The system achieved 95% accuracy in lesion detection and 98% in BI-RADS classification, optimizing medical imaging analysis with AI-driven automation.
MiceAI (Kutahya Health Science University)
The MiceAI project is an AI-assisted software for rat behavior experiments, actively used in a university medical laboratory. It processes camera feeds with OpenCV, employs a YOLOv8-based rat detection model, and features a PyQt5 GUI for real-time monitoring.
YOLO11N Traffic Sign Vision
The YOLO11N Traffic Sign Vision project focuses on detecting traffic signs using the YOLO11N architecture. The repository includes training logs, model weights, and sample prediction images, providing a comprehensive overview of the model's performance.
Ear Analysis with Hybrid Model
The Ear Analysis with CNN project focuses on gender, identity, and age classification using ear photographs. A hybrid ResNet-CNN model was developed, combining ResNet feature extractors with custom CNN layers. The system was trained on 40 different models, achieving 91% accuracy.