Arbetsbeskrivning

Transport is at the core of modern society. Imagine using your expertise to shape sustainable transport and infrastructure solutions for the future. If you seek to make a difference on a global scale, working with next-gen technologies and the sharpest collaborative teams, then we could be a perfect match.

At Volvo Autonomous Solutions (V. A.S), we drive the future of transport by developing and operating cutting-edge autonomous solutions. As part of the Volvo Group, we are committed to creating a safer, more sustainable, and highly productive transport industry—guided by our ambitions of being 100% Safe, 100% Fossil-Free, and 100% More Productive.

Our foundation is built on safety, passion, collaboration, respect, and trust. We foster an inclusive and diverse workplace where innovation thrives, and every voice matters. Whether you're shaping technology, optimizing operations, or redefining mobility, you’ll find a place here. Join us in transforming the way the world moves.

We are looking for an AI/MLFramework Engineer to join our team and help shape the future of autonomous transport systems.

About the Role

As a Framework Engineer at Volvo Autonomous Solutions, you’ll help shape the future of autonomous transport by enabling high-performance AI on cutting-edge hardware. Your focus will be neural network optimization, model acceleration, and deployment on embedded and edge platforms, ensuring our models run efficiently across GPUs, automotive-grade SoCs, and distributed systems.

You’ll design and maintain frameworks that support the full AI lifecycle—from training and optimization to real-time inference. Working closely with AI engineers, MLOps, perception teams, and hardware experts, you’ll ensure models meet strict performance, power, and latency requirements.

Your work will include optimizing deep learning models for platforms like NVIDIA GPUs, Jetson/Orin, TensorRT, ONNX Runtime, and TVM, building pipelines for graph and compiler-level improvements. You’ll also extend internal AI runtimes, develop hardware-aware libraries, and integrate models into ROS/ROS2, NVIDIA Isaac, and our autonomous driving stack.

Collaboration is key—you’ll validate compute and sensor constraints, evaluate new hardware, and build CI/CD tools for benchmarking and regression testing. Finally, you’ll lead performance validation, implementing automated tools to measure inference speed, accuracy, and resource usage, driving continuous improvements in efficiency and reliability.

What We’re Looking For

Required Qualifications

• 2+ years in deep learning frameworks, model optimization, or embedded AI deployment.
• Strong proficiency in PyTorch, TensorFlow, ONNX, TensorRT, or similar.
• Expertise in C++, Python, and hardware-aware programming.
• Experience with model compression (quantization, pruning, distillation).
• Hands-on experience with real-time inference on embedded compute (Jetson/Orin, automotive SoCs).

Preferred Skills

• Familiarity with DeepStream, TVM, MLIR, XLA, CUDA, or compiler toolchains.
• Understanding of camera/LiDAR pipeline constraints and sensor synchronization.
• Knowledge of Linux kernel performance optimizations and autonomous systems safety requirements.

Who are you?

You are a strong communicator who thrives in cross-functional environments, collaborating seamlessly with AI, robotics, and hardware teams. You bring structure and clarity to complex projects and have the ability to take full ownership of technical frameworks from start to finish.

Why Join Us?

Joining Volvo Autonomous Solutions means becoming part of a pioneering team that is redefining how the world moves. You’ll work on cutting-edge technology in a mission-driven, global environment where your contributions have a direct impact on future mobility. We offer a flexible, inclusive workplace culture that values collaboration, personal development, and innovation. Here, you’ll find the opportunity to take ownership, drive change, and grow in a company that believes in balancing purpose with performance.