Stage De Recherche En Ia: «architecture De Déploiement Next-gen : Executorch, Tvm Unity Et Accélération Hybride» (h/f)
Valeo Service
Créteil, France
Hybrid
Python and c++ mastery
Pytorch 2.x frontend expertise
Compiler architecture understanding
Valeo Service is seeking a research intern for a position focused on next-gen deployment architecture for AI in embedded systems. The role requires strong skills in Python and C++, along with a background in compilers, embedded systems, or AI, and offers an opportunity to work within a leading global tech company
Job Summary
The role involves evaluating and implementing new compilation standards like ExecuTorch and TVM Unity to bridge PyTorch models with hardware accelerators.
You will work within the Valeo Brain division's ACE team to automate the transition from research code to executable binaries on embedded chips.
This is a 5-6 month research internship opportunity focused on state-of-the-art deployment technologies in the early adoption phase.
Matching Summary
Match Score: 85
Valeo Service is seeking a research intern for a position focused on next-gen deployment architecture for AI in embedded systems. The role requires strong skills in Python and C++, along with a background in compilers, embedded systems, or AI, and offers an opportunity to work within a leading global tech company.
Skills & Requirements
Must-have
Python and C++ mastery
PyTorch 2.x frontend expertise
Compiler architecture understanding
Embedded AI system knowledge
Linux and Git proficiency
Nice-to-have
Experience with ExecuTorch or TVM
Knowledge of QNN or TensorRT SDKs
Strong abstraction and exploration skills
Fluent English technical communication
Key Requirements
Engineering degree (Bac+5) or Master 2 Research
Specialization in Computer Science, Compilers, Embedded Systems, or AI
Current student status eligible for a 5-6 month internship