Skip to content
View zmerpez's full-sized avatar

Block or report zmerpez

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
zmerpez/README.md

Zeliha Ural Merpez

Data Scientist
📧 [email protected] | 📞 +1 (778) 939-7183
🔗 LinkedIn | GitHub


💼 Professional Summary

Data Scientist with a strong foundation in machine learning, statistical modeling, and data engineering, combined with a PhD-level research background in applied mathematics and finance. Proven experience designing production-grade ML pipelines, evaluating generative AI models, and applying NLP and time-series techniques to real-world datasets. Adept at bridging theoretical concepts with practical implementation, with a track record of contributing to AI product development, academic research, and data-centric education. Passionate about building robust, interpretable, and scalable solutions that drive intelligent decision-making.


🛠 Skills

Machine Learning & AI
Supervised Learning · Regression & Classification · Computer Vision · NLP · LLM Evaluation · Generative AI · Prompt Engineering · Model Interpretability · Model Optimization · Transfer Learning

Data Science & Analytics
Statistical Analysis · Time-Series Forecasting · Data Cleaning & Preprocessing · EDA · Feature Engineering · Hypothesis Testing · A/B Testing · Data Visualization (Matplotlib, Seaborn, Plotly)

Programming & Tools
Python (Pandas, NumPy, Scikit-learn, PyTorch, TensorFlow, OpenCV, NLTK, SpaCy) · SQL · PostgreSQL · pgAdmin · Jupyter Notebook · Git · VS Code

Pipelines & Engineering
ML Pipelines · Model Deployment · ETL · Data Integration · Workflow Automation

Other
Research Design · Academic Writing · Curriculum Development · Technical Mentoring


💼 Experience

Data Science Expert

Outlier (Freelance · Remote)
Apr 2025 – Present

  • Designed and answered advanced data science questions to support AI model training with high-quality, domain-specific inputs.
  • Assessed AI-generated responses for accuracy, depth, and contextual relevance.
  • Explored ML algorithms’ theoretical foundations to evaluate model behavior.
  • Applied mathematical expertise to improve conceptual robustness of AI systems.

Machine Learning Intern

M2M Tech
Dec 2024 – Present

  • Built and deployed end-to-end ML pipelines for image regression in production.
  • Integrated models into factory systems, driving automation and operational insights.
  • Optimized models for performance and accuracy using modern tuning methods.

AI Development Specialist (Math & Code Analyst)

Outlier
Mar 2024 – Present

  • Evaluated and optimized generative AI outputs in mathematics and programming.
  • Crafted challenging prompts and test cases for robustness testing.
  • Provided feedback to improve mathematical accuracy and model alignment.

Postdoctoral Researcher – Data Science & Finance

University of British Columbia (UBC)
Sep 2021 – Oct 2023

  • Led research at the intersection of data science and finance, using predictive modeling and NLP.
  • Performed large-scale data cleaning, analysis, and visualization.
  • Built time-series models and explored financial applications of LLMs.
  • Extracted and manipulated structured data using SQL.

Instructor – Database Management

UBC (Sauder School of Business)
Sep 2021 – Dec 2022

  • Designed and taught courses on SQL, relational databases, and schema design.
  • Delivered hands-on training in PostgreSQL and SQL via Jupyter Notebooks & PgAdmin.
  • Focused on real-world application and performance optimization in queries.

📚 Education

  • Master of Data Science ( University of British Columbia)

  • Ph.D. Mathematics ( Bilkent University)

  • B.Sc Mathematics ( Bilkent University)

Pinned Loading

  1. neurologx neurologx Public

    **NeuroLogX** is an intelligent log analysis platform that stores and indexes logs using **Apache Lucene**, then applies **deep learning** models (CNNs, Transformers) to classify, detect anomalies,…

    Python 2

  2. StockVision StockVision Public

    StockVision uses machine learning to predict stock price movements based on historical data. By leveraging algorithms like time-series analysis and neural networks, it provides insights into future…

    Jupyter Notebook 1 1

  3. dark-RL dark-RL Public

    Unleashing Strategy in a Puzzle Game with Reinforcement Learning

    Jupyter Notebook

  4. simple_jobs simple_jobs Public

    Data wrangling, cleaning

    Jupyter Notebook

  5. UBC-MDS/Workflows_showcase UBC-MDS/Workflows_showcase Public

    Group repository for group 25 in DSCI 522

    Jupyter Notebook 2 6

  6. UBC-MDS/DSCI525_Group14 UBC-MDS/DSCI525_Group14 Public

    Web and Cloud Computing

    HTML 1 1