Matthieu
Olekhnovitch

Description

As a Quant Researcher Intern at Société Générale, I worked on mid-frequency trading strategies, specifically on the development of predictive models for volatility and returns. My main focus was on most recent academic ML approaches and implementing them in a real-world trading environment. Topics included: volatility prediction, predictability, factor models, and more.

Technologies Used

• Pandas
• PyTorch

Supervisors

• Dounia Ouacha
• Selmen Jnifen

Description

Founded laive.ai, an API-first platform that empowers developers to create sophisticated AI agents with advanced knowledge processing abilities. As technical co-founder, I architected a comprehensive ML infrastructure that provides developers with a full stack of endpoints for document understanding, contextual retrieval, and intelligent task execution. Built core features including real-time semantic extraction with both dense and sparse vector search, automated knowledge graph construction from diverse data sources, and context-aware information synthesis. Engineered a highly scalable architecture able to process 1M+ API calls daily, with dedicated ML pipelines for continuous model improvement. The platform supports various agent frameworks and LLM integrations (Python SDK, Langchain, LlamaIndex, etc.). Implemented authentication, rate limiting, and monitoring systems while maintaining availability. Set up Devops and CI/CD pipelines.

Technologies Used

• System Design
• Neural Language Processing
• Retrieval Augmentation
• DevOps

Description

Conducted research with the Serval PhD team in automated data augmentation for time series analysis. My work focused on extending the 2020 AutoAugment algorithm, build for image data, to handle the complexity and high-dimensionality of time series data across various domains including medical diagnostics, financial markets, and industrial IoT sensors. I developed novel VAE-based methods that could automatically learn optimal augmentation policies for different time series characteristics, significantly improving model generalization capabilities and beating all state-of-the-art results. The research involved implementing custom PyTorch modules with optimization libraries such as Pyro for complex distribution modelling. My findings were compiled into a research paper submitted to ICML 2026, demonstrating substantial improvements in classification accuracy across multiple benchmark datasets.

Technologies Used

• Pandas
• PyTorch
• Scipy
• Weight & Biases

Supervisors

• Prof. Sylvain Kubler
• Prof. Yves Le Traon

Description

At Worldline, I worked on the development of an intelligent code analysis system that leveraged early RAG techniques to understand and analyze large-scale software projects. I implemented transformer-based encoder models for sentiment analysis of code comments, commit messages, and documentation, enabling the system to detect and predict potential issues during merges. The system featured early RAG (Retrieval-Augmented Generation) pipelines that could intelligently search through codebases, documentation, and historical data to provide contextual insights to development teams. The final system was entirely language agnostic, allowing it to be used with any programming language and framework and across projects with multiple languages. My tutor was Marie-Alice Blete, later wrote a pioneering book on RAG and agentic techniques for large scale applications.

Technologies Used

• RAG
• Langchain
• Text-encoding
• Redis

Supervisors

• Marie-Alice Blete

Description

As a Battalion Chief Officer at the Brigade de Sapeurs-Pompiers de Paris, I was responsible for overseeing emergency response operations and managing a base team of 3-4 firefighters. My role involved coordinating responses to medical emergencies, car crashes, and other incidents, ensuring the safety of both the public and my team. I also played a key role in developing training programs for new recruits, focusing on both analytical skills and leadership development.

Supervisors

• Capitaine Lemur