World-Model Scientist Intern - 3D Vision & Geometry

AMI Labs; Paris; Sur place; EUR 80 000 - 140 000; Plein temps

Résumé du poste

AMI Labs in Paris is seeking a world-model research scientist focused on 3D geometry. You will collaborate with a team of scientists and engineers to push the state-of-the-art in world models derived from video and sensor data.

Responsibilities include designing self-supervised learning methods, developing architectures that predict world dynamics from high-dimensional signals, and building scalable data processing and evaluation pipelines for planning and reasoning.

Qualifications

• Bachelor’s degree or equivalent experience in Computer Science or related field.
• Proficiency in Python.
• Ability to design, run, and analyze experiments independently.
• Understanding of machine learning fundamentals, large-scale training, and accelerator-based compute environments.

Responsabilités

• Work on world model research focusing on 3D geometry of the real world.
• Design self-supervised learning methods to learn from video and high-dimensional signals.
• Develop architectures that predict world dynamics from video and other signals.
• Create scalable algorithms for pre-processing and curating video data.
• Build evaluations for benchmarking world model understanding, prediction, and planning.
• Develop efficient algorithms for model-based planning and reasoning.

Connaissances

Python

Experimentation

Independent work

Machine learning fundamentals

Formation

Bachelor’s degree or equivalent in CS or related field

Outils

PyTorch

JAX

Blender

Description du poste

AMI Labs in Paris is seeking a world-model research scientist focused on 3D geometry. You will collaborate with a team of scientists and engineers to push the state-of-the-art in world models derived from video and sensor data.

Responsibilities include designing self-supervised learning methods, developing architectures that predict world dynamics from high-dimensional signals, and building scalable data processing and evaluation pipelines for planning and reasoning.