Skip to content
View umutluprens's full-sized avatar

Block or report umutluprens

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. 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
umutluprens/README.md

Hi, I'm Umut Yasar

Financial risk analyst with 10+ years in international banking. I work on the Financial Risk Team at AIB Group, covering capital adequacy, risk-weighted asset oversight and risk-adjusted return analysis. I hold an MSc in Financial Data Science from University College Dublin.

This is where I keep the analytical side of my work. Most projects apply Python, SQL and Julia to questions I meet in banking, such as how sound a bank is, how credit risk can be predicted and how a portfolio should be built.

I am looking for roles that bring banking and data together, in development and impact finance, trade finance or corporate and risk, and I am open to opportunities in the Netherlands.

Projects

bank-soundness-analytics
Soundness profiling of European banks using CAMELS indicators, adapted from my MSc group project at UCD, which covered 8,033 bank-year observations across 18 countries from 2010 to 2021. It includes a leakage-aware screening model, PCA and K-means segmentation, and a synthetic demo mode, since the original financial data is licensed. In the original project my part was the data collection, the CAMELS feature work and the modelling and evaluation.

credit-default-risk-modelling
A probability of default model on the public Give Me Some Credit dataset. It compares logistic regression with gradient boosting and evaluates both discrimination and calibration with ROC-AUC, average precision and the Brier score.

machine-learning-for-finance
Machine learning methods in Python (regression, classification, resampling, tree-based methods and SVM) applied to finance datasets such as credit default and market direction. MSc coursework, kept as a readable record of methods I can implement myself.

financial-theory-julia
Portfolio and financial theory in Julia, covering mean-variance optimisation, CAPM and utility theory with JuMP and Ipopt. MSc coursework.

FotoLiman (personal project)
A macOS and Windows app I built to back up photos and videos from Google Photos, phones and camera cards to an external disk. It verifies every copy and supports pause and resume. The source code is private, but the latest version is public and can be downloaded from the project site.

Experience

AIB Group, Capital & Recovery (2024 to present)
Second-line capital risk work across ICAAP, risk-weighted asset oversight, risk-adjusted return on capital analysis and regulatory reporting.

Türk Eximbank, Foreign Financial Institutions (2017 to 2024)
Credit analysis of international banks for buyer's credit exposure, KYC and AML due diligence, macroeconomic risk monitoring and correspondent banking relationships. I led the business side of an interdepartmental project that delivered a web-based bank analysis, rating and reporting tool. I defined the assessment methodology and requirements and worked with the development team. The tool cut the time to evaluate a bank from 8 days to 5.

Tools

Programming: Python (pandas, scikit-learn, statsmodels, XGBoost), SQL, Julia (JuMP), VBA
Reporting and data: Excel, Power BI, SQLite
Market data: Bloomberg, Refinitiv Eikon, S&P Capital IQ

Education and certification

  • MSc Financial Data Science, University College Dublin (EU Jean Monnet and Smurfit Scholar)
  • MBA, Istanbul Bilgi University
  • BSc Economics, Gazi University
  • Trade Credit Insurance certification, Offenburg University and IFTI

Get in touch

Happy to connect on LinkedIn.

Pinned Loading

  1. bank-soundness-analytics bank-soundness-analytics Public

    European bank soundness profiling with CAMELS indicators, risk segmentation, and explainable machine learning.

    Jupyter Notebook

  2. credit-default-risk-modelling credit-default-risk-modelling Public

    End-to-end credit default risk modelling with robust validation, calibration, and interpretable machine learning.

    Python

  3. financial-theory-julia financial-theory-julia Public

    Portfolio and financial theory in Julia. Mean-variance optimisation, CAPM and utility theory with JuMP and Ipopt. MSc coursework, UCD.

    Jupyter Notebook

  4. machine-learning-for-finance machine-learning-for-finance Public

    Machine learning methods in Python (regression, classification, resampling, trees, SVM) applied to finance data. MSc coursework, UCD.

    Jupyter Notebook

  5. fotoliman fotoliman Public

    Website and public releases of FotoLiman, an app that backs up photos and videos to your own disk on Mac and Windows. FotoLiman uygulamasının sitesi ve sürümleri.

    HTML