Maxime Bonnesoeur
Just a nerd on the run.
I build AI systems with AI — from LLM pipelines to a toilet finder for Switzerland. Currently building fintech infrastructure at Synthex in Zürich.
About
Machine Learning engineer who got hooked on NLP, computer vision, and finding the hidden gems in data. I've spent the last 7+ years shipping AI — from leading a data science team to architecting agentic systems for fintech. EPFL-trained, Zürich-based, and always looking for the next interesting problem to nerd out on.
Python . PyTorch . Hugging Face . AWS . Azure . Spark . RAG . LLM Pipelines . Cursor 💜
What I've been up to
Synthex
NOWAI Software Engineer — Founding Team
Building AI infrastructure for fintech — agentic systems that automate backoffice ops, extract financial data, and handle regulatory compliance. The kind of stuff where getting it wrong means real consequences.
42Matters / Similarweb
Data Scientist — Team Lead
Led the data science team and shipped 4 daily-operational products. The highlight: an IAB app classification system that doubled coverage and boosted accuracy by 25%. Also built graph-based app matching with Spark + CV, and integrated RAG search before it was cool.
ICRA 2020
PUBLISHEDForce Adaptation in Contact Tasks with Dynamical Systems
My first-ever published paper — designed a force adaptation system for collaborative robots. No extra sensors needed, 90% reduction in perturbations. Robotics meets reinforcement learning at ICRA in Paris.
The Trail
A career elevation profile — click any peak to explore.
Education
Bachelor of Micro-engineering
EPFL · Lausanne, Switzerland
Learned programming and the basics of micro-engineering and mechanical engineering to build complex products.
Master of Robotics — Deep Learning
EPFL · Lausanne, Switzerland
Master oriented toward machine-learning techniques applied to autonomous systems and continuous improvement workflows. Thesis: Monocular 3D vehicle detection and distance estimation with self-attention mechanism.
Professional Experience
Junior Machine Learning Engineer
Visium · Lausanne, Switzerland
- ▲ Created a NLP model to identify the preferred incentives of the Swiss political class toward predetermined domains
- ▲ Creation of a web application for automatic detection of false watches based on vibration signature analysis
- ▲ Development of orchestration pipelines via Airflow to setup VMs (Terraform), Slack channels, GitHub projects, and hour tracking
- ▲ Designed and implemented the sales tracking architecture using AWS, Snowflake, and DBT — lowered processing costs by 5x
Data Scientist
42Matters (acquired by Similarweb) · Zürich, Switzerland
- ▲ Led and developed the data science team, empowering and mentoring members. Spearheaded project formulation and execution, ensuring successful ML model deployment
- ▲ App Categorization: Advanced IAB classification with text embedding and cosine similarity — doubled coverage and increased accuracy by 25%
- ▲ Revenue Estimation: Sophisticated models to estimate app earnings for better-targeted marketing strategies and informed business decisions
- ▲ App Matching: Graph-based system using Spark and computer vision for efficient cross-marketplace app matching
- ▲ Demographics Estimation: Implemented demographic estimation through computer vision and Spark for targeted marketing campaigns
- ▲ Integrated vector databases and RAG search for easy review analysis and app discovery
Data Scientist
Similarweb · Zürich, Switzerland
- ▲ Transition and Integration Roles: Acted in transition, transitioning predictive systems from 42Matters to be integrated into Similarweb's future products
- ▲ Post-acquisition, responsible for maintaining and teaching existing data science systems to the wider teams at Similarweb
- ▲ Re-implemented and enhanced current systems to leverage advanced predictive capabilities
AI Software Engineer — Founding Team
Synthex · Zürich, Switzerland
- ▲ Built and automated backoffice operations for fintech industry, streamlining processes and reducing manual work
- ▲ Developed custom data extraction tools to aggregate complex financial data for regulatory compliance reporting
- ▲ Engineered text and graph extraction systems to automate document data retrieval
- ▲ Architected and deployed secure infrastructure stack for platform hosting and LLM deployment, ensuring regulatory compliance via IAC
Projects
Things I've built, published, and experimented with.
Let's Connect
Whether you want to discuss ML, collaborate on a project, or just chat about trail running around Uetliberg — I'd love to hear from you.
maxime.bonn@gmail.com
Book a Call
15 or 30 min — pick a time on Cal.com
Buy Me a Coffee
Fuel the caffeine addiction
Or find me on the trails around Uetliberg.
▲