Portrait of Constantin Riff

Constantin Riff

Mechanical Engineering graduate from ETH Zurich, specialised in Machine Learning and Robotics. Passionate about applying AI to real-world physical systems.

About

I hold a Bachelor's degree in Mechanical Engineering from EPFL (2019–2022), including an exchange year at La Sapienza in Rome, and a Master's degree in Mechanical Engineering from ETH Zurich (2022–2025), specialising in Machine Learning, Computer Vision and Robotics. My Master's thesis, carried out at the CREATE Lab at EPFL under the supervision of Prof. Josie Hughes and Prof. Marco Hutter, focused on leveraging large language models for physical reasoning and control of robotic arms. Alongside my studies, I completed an R&D internship at Flexis (Renault–Volvo joint venture, Paris), working on PLM/3DEXPERIENCE environments, a COβ‚‚ assessment framework and internal AI adoption. I am currently looking for my first full-time position at the intersection of engineering and artificial intelligence. Outside of work, I enjoy hiking, ski touring, mountaineering and playing the piano.

Projects & Publications

Master's Thesis β€” Leveraging Emerging Physical Knowledge in LLMs Toward World-Aware Robotics
CREATE Lab, EPFL / RSL, ETH Zurich β€” 2025 β€” supervised by Prof. Josie Hughes and Prof. Marco Hutter

Perception–reasoning–action framework using an LLM for physical prediction and control of a UR3 robotic arm, without task-specific fine-tuning.

Semester Project at LBB β€” Microfluidics for at-home blood testing
ETH Zurich, Laboratory of Biosensors and Bioelectronics β€” in collaboration with Hemetron

Design, prototyping and testing of a microfluidic device for at-home blood analysis.

AirPods β€” Sustainability Analysis
ETH Zurich, Advanced Manufacturing Lab β€” course project β€” supervised by Dr. LΓ©a Deillon

Life cycle analysis and environmental footprint of AirPods.

Mechanical Design Project β€” Manual Seed Drill
EPFL β€” 1st year Bachelor

Complete design of a manual seed drill in CATIA and Fusion 360, respecting tolerances and manufacturing constraints, with FEM analysis of the digital twin.

Act for Change Lab β€” 1st place: Global Energy Challenges
EPFL β€” 2020

Team case study on data centre urbanisation and energy recovery.