Luis Arean

**Context**: France's largest third-generation synchrotron, SOLEIL. Dozens of experiments on various beamlines (visible, X-ray, ultraviolet, etc.).
**Problem**: Experiments on the beamline required manual configuration changes of the monochromator, preventing automation and forcing researchers to constantly monitor the experiment, even at night.
**Role**: Senior Consultant / Tech Lead. Defined application architecture within the constraints of existing drivers. Technology choices, application development.
**Result**: Application delivered to production, stable and performant. Team autonomous for future developments.
**Stack**: C++, MicroManager (proprietary tool), Windows

**Context**: A highly secure environment for accessing the supercomputer. "Batch" culture (supercomputer jobs are always batched). Constant project flow with changing needs.
**Problem**: A legacy application with a minimal web interface and over 30 batch processes for updating the database. No immediate, online updates. Few online tools for the end-user. Unnecessary replicated databases. Obsolete language: Perl.
**Role**: Senior Expert Consultant / Tech Lead. Defined architecture, technology choices, structured the Python backend and frontend, implemented development methodologies and a non-existent DevOps environment. Developed 90% of the code. Automated tests using pytest. Configurable workflow in the backend.
**Result**: Application delivered to production, performant. Reduced workflow approval times from several days to less than 24 hours. Clean, maintainable, and well-documented database. Team autonomous for future developments.
**Stack**: Python Django RDF PostgresSQL Docker Apache Ubuntu
**Other Technologies**: RabbitMQ, LDAP, OAuth2

**Context**: The C2RMF (Louvre) particle accelerator is used by researchers to analyze artworks from a materials perspective.
**Problem**: The detector is suitable for 2D surfaces because the accelerator beam must be perpendicular to the surface being analyzed. However, for statues or 3D art objects, there's an orthogonality issue due to irregular surfaces that don't always align with the beam. The solution must be economical and fully automated (no manual repositioning each time). The architecture must align with the existing architecture of the accelerator's applications.
**Role**: Senior Expert Consultant / Tech Lead and Project Manager. Defined technical architecture, technology choices, selected positioner components through public tender. Developed the control application with a graphical interface.
**Result**: Selected a hexapod positioner capable of positioning most objects more economically than a robotic arm (tens of thousands of euros versus millions). Developed an application with an interface to a scanner that digitizes a map of the object under analysis. Used rotation models based on Euler angles to make the hexapod follow the map. Application delivered to production, performant, and economical. Automation now allows 24/7 use of the accelerator, optimizing its technical lifespan.
**Stack**: Proprietary tools on Windows, particularly LabView, which interfaces with the rest of the New Aglae architecture.