Data Science professional with strong analytical, machine learning and financial modeling skills.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Highly motivated Data Science professional with a BSc in Politics and Data Science, specializing in machine learning, financial analytics, and NLP. Proven ability to build predictive models, analyze complex datasets, and deliver actionable insights through projects like ATP Tennis Match Outcome Prediction and Equity Report on Ametek Inc. Strong technical skills in Python, R, SQL, and various ML frameworks, complemented by experience in market segmentation and sales support.
London School of Economics
BSc · Politics and Data Science
August 1, 2023 – June 30, 2026
Collège du Léman
International Baccalaureate
June 1, 2014 – May 31, 2022
DNA Payments
Summer Intern
July 1, 2021 – September 1, 2021
London, England, United Kingdom
Equity Report: Ametek Inc.
March 1, 2026 – Present
Collaborated on a full equity research report on Ametek Inc., leading the earnings quality analysis. Estimated discretionary accruals and accruals quality via cross-sectional OLS regression on large-scale financial panel data, constructing composite earnings quality scores across peer comparables. Conducted sensitivity analysis on DCF valuation model, varying revenue growth and WACC based on bull/bear forecast scenarios.
ATP Tennis Match Outcome Prediction
January 1, 2026 – Present
Built an ATP match outcome prediction pipeline, engineering 123 temporal features across player skill, form, and surface dynamics. Trained XGBoost and MLP classifiers on 58,000 ATP matches, tuning via Optuna with regularisation methods to achieve 67.2% test accuracy. Evaluated Model calibration, finding MLPs are systematically less calibrated than XGBoost, and assessed implications for threshold-hedged betting profitability.
Predicting Radical Right Support: Socio-Economic Determinants
May 1, 2025 – Present
Merged ESS survey data with PopuList 3.0 data to construct a cross-national dataset of 10,283 points and 46 features, and trained a CatBoost classifier with F1-optimised Thresholds. Applied SHAP analysis, revealing cultural and educational factors outperform economic dissatisfaction as predictors. Further national-level analysis revealed heterogeneous drivers across countries.
NLP Analysis of Financial Consumer Complaints
April 1, 2025 – Present
Applied LSA topic modelling to CFPB complaints using TF-IDF to surface latent consumer issues. Leveraged dimensionality reduction techniques to cluster companies by complaint profiles using DBSCAN and LOF, identifying complaint profile clusters and problematic companies.
Runa Capital: Venture Capital Database
January 1, 2025 – Present
Built a MongoDB venture capital database for Runa Capital, consolidating data across CrunchBase, Dealroom and LinkedIn with 150+ features per company. Implemented queries covering P/E undervaluation screening, time series analysis, and vector similarity searches.
Cultural Fit Analysis
The candidate's academic projects demonstrate a strong interest in diverse applications of data science, from financial analysis to social science and sports analytics. This breadth of interest suggests adaptability and a curious mindset, which can contribute positively to a dynamic team culture. The 'Activities & Interests' section also indicates involvement in team sports, implying teamwork and leadership qualities. However, the lack of extensive professional experience makes it difficult to fully assess cultural fit in a corporate setting.
Soft Skills & Operational Fit
The candidate's project descriptions indicate an ability to work collaboratively (Equity Report) and independently on complex data science problems. The academic background in Politics and Data Science suggests an interdisciplinary approach, which can be valuable for understanding real-world problem contexts. However, the limited professional experience means operational fit in a corporate data science environment is largely unproven.