Hey, I’m David, I'm a uni student that builds projects blending economics and machine learning

About Me

  • Me in 50 words

    I'm a 2nd year Economics student at the University of Exeter with a passion for not only my subject, but also Data-Science, Machine Learning and Mathematics.

    I don’t just like theory, I care about the applications of these tools to real world problems so I’m constantly making projects.

    Areas of Interest

    risk modelling, energy economics, computational economics.

  • What sets me apart

    1. Genuine Enthusiasm- I love this work and learn fast because I can’t stop thinking about it

    2. Deep focus and planning- I work in long, structured sprints with clear goals and a disciplined process.

    3. Fast Learning- In just one year, I’ve built a portfolio of projects that I believe speak for themselves, all on top of maintaining a 1:1 grade average.

  • Where I’m at now I’m delivering above my level, building well-documented projects with real-world relevance.

    Where I could be I learn fast, work with discipline, and approach problems methodically. With the right mentorship, I’ll scale quickly and become a high-impact contributor faster than most early-stage candidates.

  • Whilst I do love data, I’m not a machine, there are other things in life I enjoy…

    Hobbies and Interests

    Cooking: I really like cooking, I find it relaxing and I like trying foods from other cultures

    Reading: I like to read sci-fi books, currently working through Dune, personal favourites are Liu Cixin’s “Three Body Problem” series.

    Games: Ashamed to say it but I play League — I hope this doesn’t affect my job prospects too much.. I also like strategy/resource management games big fan Civilization 5, Factorio, Cities : Skylines, etc.

Technical Skills

Core Programming skills

Python (functions, modules, error handling, data structures) Web scraping (requests, BeautifulSoup, Selenium) Scripting & automation (task scheduling, CLI tools) Relational Databases ( SQL querying, joins, aggregations)

Data Science and Machine Learning skills

Data Cleaning (Pandas and NumPy) Data Visualisation (Matplotlib, Seaborn)

End-end model development (Scikit-Learn and Tensorflow) Extract Transform Load Pipelines Statistical Testing (T-Tests, ANOVA, Cohen’s D, Chi Square)

Tools, Systems and Deployment

Docker & virtual environments Version control (Git, GitHub), Linux CLI (Bash) Dash/Streamlit (interactive dashboards, data apps) Linux Distribution (Pyinstaller, AppImages) Simple GUIs (Tkinter)

this is me 🡵

You can also find me on github and LinkedIn

My Recent Projects

  • Forecasting Tail risk with ML (currently working on)

    Predicting CVAR with LSTM models

    Built an LSTM-based model to forecast short-term portfolio returns and generate forward-looking CVaR estimates.


    Combined these forecasts with a genetic algorithm to optimize portfolios for return while constraining predicted tail risk — rethinking traditional, backward-looking risk metrics.

    Key Links

  • Auto Connect (desktop App)

    Automating LinkedIn

    I built a standalone Linux desktop app that automates a data-driven outreach strategy on a popular professional networking platform. Using Python, I developed the full pipeline from scratch incorporating smart decison making with Machine Learning.

    I wrapped the entire system into a clean desktop GUI using Tkinter, bundled it with PyInstaller, and made it runnable as a standalone .exe, and distribtuing as a Linux AppImage

    Key Links

    1.GitHub 2. v1 release

  • Green Energy Adoption game

    Game Theory Meets Climate Policy

    A dynamic game theory model simulating global renewable adoption as a threshold public goods game. Nations weigh evolving costs, climate urgency, and pressure over time.


    I simulate the evolution of this in Python and then deployed as a web app to visualise how early adoption or free-riding shapes global outcomes.

    Key Links

    1. web-app 2.github 3.model description

  • Banks and macro-volatiltiy

    Banking Structure & Macroeconomic Volatility

    Explores how banking market structure relates to macroeconomic stability, this analysis applies K-Means clustering to nations on key banking and financial inclusion metrics.

    I analyse these clusters and perform statistical tests to prove a link between banking market competitiveness and lower inflation volatility

    Key Links

    github

  • Credit Default Prediction

    My First Steps in ML…

    My first ML project, focused on predicting loan defaults using application and credit card data.


    I documented every step in extreme detail — from pipelines to feature selection — gaining a deep understanding of ML workflows and best practices.

    Key Links

    github