Catalin Bondari QFA

Catalin Bondari QFA · Dublin, Ireland

Investment data & reporting

I work on the data portfolio managers rely on, and build tools in SQL, Python and Snowflake to make it faster and more reliable.

View work Get in touch Download CV
Saved by a regulatory attestations systems integration
€55k
Risk & Compliance Graduate Programme
2 yrs
Qualified Financial Adviser, Institute of Banking
QFA 2025
BSc Business Studies with Data Analytics, DCU
2.1 Hons
  • UCITS
  • VaR
  • Snowflake
  • SQL
  • Python
  • Claude
  • Jacobi
  • FactSet
  • Bloomberg
  • Power BI
  • Power Automate
  • Excel (Power Query)
  • Investment data quality
  • Performance & risk reporting
  • Regulatory reporting
  • AI-assisted workflows

About

I work on the investment data that portfolio managers rely on every day. As part of Davy’s Portfolio Insights team, I help maintain the quality of holdings, performance, exposure and risk data across our UCITS fund range and segregated portfolios. That data feeds client reporting, board and ManCo packs, and SAA reviews. Day to day I work in Jacobi, Snowflake, FactSet and Excel alongside our portfolio managers and quants.

I also build. I use Claude to design AI-assisted workflows that take manual validation and reporting work off the team. Pair that with SQL and Python, and checks that took hours become repeatable, auditable processes.

Before this, I spent two years on Davy’s Risk & Compliance Graduate Programme. In Financial Risk I produced daily VaR reporting and ran backtesting and stress tests for the Financial Risk Committee. In Capital Markets Compliance I covered MAR trade surveillance and CBI and ESMA regulatory submissions. One notable contribution, a systems integration for regulatory attestations, saved the firm €55k.

Experience

Four stations: the timeline, the data, the risk work and the toolkit. Each instrument responds to drag, hover or its own controls.

  1. Timeline
  2. Portfolio data
  3. Risk & compliance
  4. Toolkit

Timeline

  1. Aug 2026 – Present

    Current

    Investment Data & Reporting Specialist

    Portfolio Insights · Davy

    Quality of holdings, performance, exposure and risk data across UCITS funds and segregated portfolios.

  2. Sep 2024 – Aug 2026

    Complete

    Risk & Compliance Graduate Programme

    Davy

    1. Sep 2025 – Aug 2026
      Rotation 2 · Financial Risk

      VaR reporting, backtesting and stress testing.

    2. Sep 2024 – Sep 2025
      Rotation 1 · Capital Markets Compliance

      MAR surveillance, regulatory submissions, Power BI and Power Automate MI.

  3. Jun – Aug 2024

    Office Manager

    Apollo Language Centre

  4. Jul 2021 – Jun 2023

    Romanian–English Interpreter

    Accord Language Translation Service

Portfolio data

Synthetic · illustrative
Data
Holdings, performance, exposure, risk
Tools
Jacobi, Snowflake, FactSet, Excel

The chart is a synthetic fund price over 16 weeks, generated for this page. It is not real data.

W16 · O 101.53 · H 102.49 · L 99.91 · C 100.90

Risk & compliance

Synthetic · illustrative

A sonar sweep over the risk and compliance work. Signal positions and levels are synthetic.

Select a signal to read it

Toolkit

Data & engineering
  • SQL
  • Python
  • Snowflake
  • Excel (Power Query)
  • Power BI
  • Power Automate
Investment platforms
  • Jacobi
  • FactSet
  • Bloomberg
AI
  • Claude
  • AI-assisted workflows
Domain
  • Investment data quality
  • Performance & risk reporting
  • UCITS
  • VaR
  • Regulatory reporting
Languages
  • English
  • Romanian

Selected work

  • W-01 · Personal projectSynthetic data

    Bond Portfolio Centre

    Demo with synthetic data. Not investment advice.

    A personal portfolio analytics app: 1-day VaR, a Monte Carlo fan chart, diversification and stress tests, optimiser mixes, a rebalance plan, goal glide paths, and a rule-based investment committee that reviews each book.

    Python · pandas · vanilla JS · Chart.js

  • W-02 · Personal projectPublic data

    Weekly Market Update

    A click-through weekly markets presentation: pick a view, then step through the week’s stories with charts built from public data, plus presenter notes.

    HTML/JS · custom chart engine · Python build/validate

College work is under Projects: six DCU projects with full write-ups.

Projects

Six projects from my BSc in Business Studies with Data Analytics at Dublin City University. Each opens a full write-up: what it is, the method, the tools and how it was evaluated.

  • P-01With Bohdan Boiprav

    Phishing email classification

    Six classical models and a fine-tuned BERT transformer, compared on telling phishing emails from legitimate ones, with LIME explanations.

    Python · PyTorch · Hugging Face Transformers · LIME

  • P-02Solo project

    Telegram BI bot

    A scheduled Python bot that scrapes websites, pulls live business news from NewsAPI and stores it in PostgreSQL, then serves it in Telegram.

    Python · BeautifulSoup · requests · PostgreSQL

  • P-03Solo

    Time-series features

    Activity recognition from accelerometer data: segment the signal, extract statistical features, then classify with Random Forest and Linear SVC.

    Python · pandas · NumPy · scikit-learn

  • P-04Solo

    Daft.ie data analysis app

    Scrapes and cleans Daft.ie property listings, with login, search filters and an interactive dashboard of charts.

    Python · BeautifulSoup · Requests · Tkinter

  • P-05MT412 group project

    Power BI sales analysis

    A Power BI dashboard of Blackbird’s sales: orders and profit, maps by state, shipping methods and the effect of discounts.

    Power BI

  • P-06With Bohdan Boiprav

    Bank customer churn prediction

    Predicts which bank customers are likely to leave, using the Kaggle BankChurners data: one-hot encoding, SMOTE, PCA, then Random Forest vs SVM.

    Python · SMOTE · PCA · Random Forest · SVM

What I do

  • Data analysis & reporting

    Power BI dashboards that turn large datasets into clear charts, using a wide range of chart types to communicate the key findings.

    See: Power BI sales analysis

  • Automation

    Power Query workflows, and compliance reporting automated with Power BI and Power Automate, to find smarter ways to work with data.

  • Web scraping

    Beautiful Soup and requests to collect and clean data from the web, as in the Daft.ie data analysis app and the Telegram BI bot.

    See: Daft.ie data analysis app

  • Machine learning

    Logistic regression, decision trees, random forests, gradient boosting, naive Bayes and multi-layer perceptrons, plus a BERT transformer for email classification.

    See: Phishing email classification

Writing

Notes on machine learning, NLP, Python, BI and data work. The six most recent posts:

All 15 posts

Media

Videos I recommend: investing first, then data and AI. The video player loads from YouTube (privacy-enhanced mode) only when you press play.

Reading & listening

Books

  • The Hour Between Dog and Wolf

    A neuroscientist and former trader on how hormones and stress biology drive risk-taking on trading floors.

    John Coates

  • Advances in Financial Machine Learning

    Practical machine learning for finance: labelling, backtest overfitting and feature importance.

    Marcos López de Prado

  • Machine Learning for Asset Managers

    A short Cambridge Element on machine learning for portfolio construction, clustering and denoising covariance matrices.

    Marcos López de Prado

  • The Man Who Solved the Market

    The story of Jim Simons and the quantitative approach behind Renaissance Technologies.

    Gregory Zuckerman

  • Expected Returns

    A deep, evidence-based survey of what drives returns across asset classes and strategies.

    Antti Ilmanen

  • Hands-On Machine Learning with Scikit-Learn, Keras & TensorFlow

    A practical guide to building machine learning models in Python, from scikit-learn basics to deep learning with Keras and TensorFlow.

    Aurélien Géron

  • The Elements of Statistical Learning

    The standard statistical treatment of supervised learning: linear methods, regularisation, trees, boosting, SVMs and model assessment.

    Trevor Hastie, Robert Tibshirani and Jerome Friedman

  • Pattern Recognition and Machine Learning

    A probabilistic, Bayesian treatment of pattern recognition: graphical models, kernel methods, mixture models and approximate inference.

    Christopher Bishop

Podcasts

Education & credentials

  • QFA · Institute of Banking 2025
  • BSc Business Studies with Data Analytics · Dublin City University · 2.1 Honours 2021–2024

Skills are listed on the Toolkit station in Experience.

More

Interests

I am passionate about investments, financial markets, and the application of AI and process automation in finance. I am actively building my knowledge across equity analysis, fixed income, derivatives and portfolio management.

I am genuinely excited by the role AI and automation play in modern finance. Whether it is building Power Query workflows, automating compliance reporting with Power BI and Power Automate, or exploring how machine learning is reshaping investment decision making, I enjoy finding smarter ways to work with data.

Contact

New messageFormspree

Send a message and I’ll reply by email. All fields are required.

Elsewhere