The purpose of this project is to design and build an open-source Fourier-Transform Infrared (FTIR) spectrometer, making advanced spectroscopy more accessible to research laboratories, startups, and companies that require rapid, reliable spectral analysis. By combining open hardware with powerful open-source software (including machine learning tools for automated spectrum interpretation) our aim is to provide a flexible, affordable, and user-friendly platform for material and chemical analysis.

The initiative brings together students from diverse fields at EPFL, ranging from physics and optics to computer science, electronics, and machine learning. The team will gain hands-on experience in optical design, interferometry, signal processing, embedded systems, and data-driven modeling, while developing a tool that can be directly reused and adapted by the scientific community. Beyond building the prototype, the project seeks to establish a collaborative knowledge base: detailed documentation, reproducible designs, and modular software. This will enable external teams to replicate, improve, and deploy the spectrometer in their own contexts, thereby accelerating innovation and democratizing access to advanced spectroscopy.

The adventure starts with the first prototype, aiming to demonstrate robust measurement performance and automated spectrum classification. The long-term vision is to create a fully modular ecosystem) hardware, software, and AI (that can evolve with the needs of the community and support applications from academic research to industrial quality control.