Ayda Firouzi

My research aims to give a foundational vision to software engineering and Agentic AI, turning fragile scripts into smart systems that navigate the messy web without constantly breaking down. I love bridging practical web automation with intelligent systems.

My work directly spans software engineering, AI, explainable AI, deep learning, natural language processing, and human-computer interaction.

Multimodal ML Foundation Models Explainable AI NLP Software Testing Healthcare AI RAG Systems Data Science

Education

Istinye University Oct 2023 – Sep 2025
Master of Data Science  ·  Istanbul, Türkiye
CGPA: 3.38 / 4.00
Thesis: "Software Test Cases Prioritization using Data Analysis and Machine Learning Algorithms"
University of Tabriz Sep 2018 – Jul 2023
Bachelor of Materials and Metallurgical Engineering  ·  Tabriz, Iran
CGPA: 18.18 / 20.0
Final Project: Predicting Thermophysical Behavior of Phase Change Materials Using Machine Learning
Relevant Coursework:  Guided Research · Data Science · Machine Learning Operations · Database Applications · Deep Generative Models · Optimization Methods · Software Modeling and Design

Publications & Submissions

  • Predicting Thermophysical Behavior of Iconic Liquid-Based Fatty Acid Phase Change Materials Using Machine Learning Algorithms
    Ayda Firouzi, Mohammad Alipour
    Manuscript in preparation
    In Progress
  • Coverage-driven Software Test Case Prioritization Using Structural Feature-Oriented Machine Learning
    Ayda Firouzi, Bahman Arasteh
    Submitted to Springer, Iran Journal of Computer Science
    Under Review
  • Effectiveness of Real-Time AI-Guided Surgical Decision Support on Operative Time and Intraoperative Blood Loss: A Meta-Analysis of Controlled Clinical Studies
    Ayda Firouzi, Mahsa Asadi Anar
    Submitted to Springer, BioMedical Engineering Online
    Under Review
  • Machine Learning Algorithms for Predictive Analytics in Big Data
    Ayda Firouzi
    9th International Conference on Applied Researchers in Science and Engineering (ICAR 2026) — Oral Presentation
    Accepted

Research Experience

Marmara University
Undergraduate Researcher
Supervisor: Prof. Cengiz Bahçekapılı
  • Designed a scalable ETL pipeline scraping 4 years of residential price index data across 81 Turkish cities, reverse-engineering Cloudflare-protected REST APIs via browser DevTools and Selenium-based session extraction.
  • Processed and normalised raw JSON outputs into multi-currency Excel exports; built Tableau dashboards for comparative year-over-year price trend analysis.
  • Designed and trained a neural network for one-year-ahead residential price forecasting.
Istinye University
Undergraduate Researcher
Supervisor: Assoc. Prof. Bahman Arasteh
  • Led a metaheuristic test data generation project, implementing algorithms across benchmark programs and building a full-stack Vue.js pipeline to automate feature extraction from HTML coverage reports — eliminating manual processing across 1,158 test instances.
  • Extracted features per test case and trained classification and regression models to predict test effectiveness and coverage quality.
Istinye University PcLab
Undergraduate Researcher
Supervisor: Assoc. Prof. Okan Yaşar
  • Designed an end-to-end ML classification pipeline to predict student dropout risk from 840+ records, applying multi-factor risk scoring across GPA, faculty, payment status, and scholarship variables.
  • Built a RAG-based querying system using Qdrant vector database and a locally deployed LLM, enabling natural language exploration of structured student data.
  • Developed a FastAPI REST backend and interactive Streamlit dashboard for real-time visualisation; containerised the full system with Docker for reproducible deployment.

Academic Experience

  • Teaching Assistant — "Large Language Models," Istinye University · Supervisor: Dr. Okan Yaşar
  • Teaching Assistant — "Machine Learning Applications in Business," Istinye University · Supervisor: Dr. Peren Jerfi
  • Volunteer Mentor — Girls Who Code: mentored high school students on Python and web development

Technical Skills

Programming & Tools
PythonC++SQL GitLinuxSelenium BeautifulSoupVue.js FastAPIDocker
ML & Deep Learning
PyTorchNumPyPandas Neural NetworksNLP Genetic AlgorithmsMatplotlib SeabornQdrant
Cloud & Platforms
AWSGCPAzure
Analytics & Other
TableauPower BI RAG SystemsWeb Scraping Agile / ScrumStreamlit

Languages

Persian
Native
Turkish
Native / Bilingual
Azerbaijani
Native / Bilingual
English
Proficient · TOEFL iBT 100
German
Beginner · A2

Contact

Email
aydafiruzy@gmail.com
Phone
(+90) 534 350 5847
Location
Istanbul, Türkiye