About me

French and Swiss second-year Master’s student in Data Science at EPFL, with a minor in Financial Engineering. I have a strong background in mathematics, programming, and applied machine learning, with interests spanning data science, quantitative finance, and LLMs. I’m particularly drawn to systematic investing, derivatives, and risk modeling. I completed an AI internship at Adequasys and currently work part-time in IT support at EPFL (Poseidon). Curious, adaptable, and detail-oriented, I enjoy solving complex problems at the intersection of technology and finance.

🎓 Education

École Polytechnique Fédérale de Lausanne (EPFL)
M.Sc. in Data Science with a Minor in Financial Engineering
📍 Lausanne, Switzerland — 2024 – Present

École Polytechnique Fédérale de Lausanne (EPFL)
B.Sc. in Communication Systems
📍 Lausanne, Switzerland — 2020 – 2024

Lycée Louis Lachenal
French Scientific Baccalauréat with Honours (Mention Très Bien)
📍 Annecy, France — 2017 – 2020

💼 Experience

Poséidon
Level 1 and 2 Support in the Poseidon IT Department at EPFL
📍 Lausanne, Switzerland — Sep. 2024 – Present

  • Provided technical assistance to students and staff to resolve IT issues.
  • Diagnosed and repaired computers, addressing both hardware and software aspects.

Adequasys
Artificial Intelligence Intern
📍 Chavannes-de-Bogis, Switzerland — Jul. 2024 – Aug. 2024

  • Design, development, and deployment of a chatbot on Teams using Copilot Studio for managing absence requests and accessing employee information.
  • Integration of Adequasys’ internal API to retrieve and process data in real-time.
  • Optimization of chatbot interactions using advanced prompting techniques and generative AI.
  • Implementation of a secure authentication system via JWT and Microsoft Entra ID.
  • Writing detailed and structured documentation to ensure effective knowledge transfer.

EPFL Artificial Intelligence Laboratory (LIA)
Teaching Assistant - Artificial Intelligence at EPFL
📍 Lausanne, Switzerland — Feb. 2024 – Jun. 2024

  • Answered students’ questions to clarify course concepts and guided them in solving exercises.
  • Participated in grading assignments and exams, as well as monitoring tests.

EPFL Laboratory for Quantum Gases (LQG)
Teaching Assistant - Mechanical Physics at EPFL
📍 Lausanne, Switzerland — Sep. 2022 – Dec. 2022

💼 My projects

  • 🧠 LiSoViMa: An Educational LLM Assistant

    📍 EPFL – Master in Data Science, Year 1 (2025)

    LiSoViMa is a modular and optimized educational language model built for STEM-related multiple-choice question answering (MCQA). The project explores supervised fine-tuning (SFT), quantization, retrieval-augmented generation (RAG), and preference-based alignment (DPO) using Qwen3-0.6B-Base. We benchmark the performance, cost-efficiency, and alignment of each variant using curated STEM datasets and preference feedback.

    GitHub Repository

    Report

  • 🌪️ VIX and Variance Derivatives: Theory, Pricing, and Replication

    📍 EPFL – Master in Financial Engineering, Year 1 (2025)

    This project explores the theory and practical implementation of VIX and variance-linked derivatives. We derive and interpret the Carr-Madan replication formula, model the dynamics of implied volatility and realized variance, and analyze the pricing of VIX futures, variance futures, and replication strategies using volatility-linked instruments.

    GitHub Repository

    Report

  • 📈 Multi-Horizon Volatility Forecasting using ODEs and Cross-Stitch Networks

    📍 EPFL – Master in Data Science, Year 1 (2024)

    We developed a hybrid neural architecture for predicting multi-horizon volatility in high-frequency trading using the FI-2020 dataset. The architecture combines ODE networks and Cross-Stitch networks to model complex temporal dynamics in financial data. Our ODE-based Cross-Stitch model outperformed the Temporal Fusion Transformer, a state-of-the-art baseline.

    GitHub Repository

    Report

  • 🪙 Crypto Carry Trade Strategies with Perpetuals, Staking and Pendle

    📍 EPFL – Master in Financial Engineering, Year 1 (2025)

    This project investigates delta-neutral crypto carry trade strategies combining perpetual futures, staking mechanisms, and tokenized yield via Pendle Finance. We analyze historical performance, market resilience, and risk profiles across major crypto events from 2019 to 2024.

    GitHub Repository

    Report

  • 📊 International Diversification & Multi-Asset Strategies

    📍 EPFL – Master in Financial Engineering, Year 1 (2025)

    This project evaluates the benefits of international diversification and dynamic asset allocation strategies, combining global equity indices and currency overlays. We implement equal-weight, risk-parity, and mean-variance portfolios, and enhance them with momentum, reversal, carry, and dollar strategies. Performance is assessed through backtests and Fama-French factor regressions from 2002 to 2024.

    GitHub Repository

    Report

  • 🎬 From Screen to Heart – How Do Emotions Shape the Cinematic Landscape?

    📍 EPFL – Master in Data Science, Year 1 (2024)

    🏆 Selected among the Top 10 Projects (out of 100+ teams)

    This project investigates the emotional dimensions of storytelling using a large corpus of over 11,000 movie summaries enriched with 450,000 IMDb reviews. We used HuggingFace’s Emotion DistilRoBERTa model to extract primary emotions (joy, sadness, fear, anger, surprise, disgust, neutrality), then applied statistical correlations, PCA, and K-Means clustering to map emotional trends across genres and narratives.

    GitHub Repository

    Demo

👉 See all my projects on the Projects page.

📬 Contact

Feel free to reach out to me for any inquiries, collaborations, or just to say hello!

Email: contact@matthiaswyss.fr
LinkedIn: linkedin.com/in/matthiaswyss
GitHub: github.com/matthias-wyss
Tel: +33782836003