Senior Compiler Software Engineer

Magics Technologies
Geel, Belgium
18 days ago

Role details

Contract type
Permanent contract
Employment type
Full-time (> 32 hours)
Working hours
Regular working hours
Languages
English
Experience level
Senior

Job location

Geel, Belgium

Tech stack

Artificial Intelligence
Artificial Neural Networks
C++
Computer Programming
Programming Tools
Logic Synthesis of Circuits
Embedded Software
Python
Real-Time Operating Systems
TensorFlow
System Programming
PyTorch
Information Technology
ONNX (Open Neural Network Exchange) Format
Bare Metal

Job description

Magics is hiring a Software Engineer to help shape the software stack for our next-generation AI accelerator. This is a technical, high-impact role focused on AI compiler development, low-level software tooling, and deep hardware-software co-design.

You'll work closely with our digital design and embedded systems teams to build the tools and frameworks that bridge advanced neural networks with custom silicon - enabling AI to operate reliably in space, energy, and defense systems., This is a full-time hybrid role for a Software Engineer, based in Leuven or Geel, Belgium, focusing on AI compilers and low-level software development.

  • Develop and extend our AI compiler toolchain, adding support for our custom accelerator and applying hardware-aware optimizations to model graphs.
  • Collaborate closely with digital chip designers to drive true hardware-software co-design for our upcoming AI accelerator.
  • Build peripheral drivers, SDKs, and developer tools that make it seamless for users to integrate their applications with our hardware solutions.
  • Create and optimize low-level libraries and kernels (e.g., convolution operators for TVM) used within our compiler stack.

Requirements

Do you have a Master's degree?, * Master's/PhD in Electrical Engineering, Computer Science, or equivalent practical experience.

  • Experience with AI compiler frameworks such as TVM or IREE.
  • Strong programming skills in Python and C/C++.
  • Deep familiarity with PyTorch, TensorFlow, ONNX, or related NN frameworks.
  • Knowledge of quantization, training pipelines, or model compression.
  • Embedded software development, including bare-metal or RTOS environments.
  • Ability to build end-to-end demonstrators, prototypes, or system-level integrations.

Benefits & conditions

  • Solve real, high-stakes engineering challenges at the intersection of AI, radiation-hardened computing, energy, and space.
  • Work shoulder-to-shoulder with top-class engineers, shaping technologies that push the boundaries of what's possible.
  • Be part of a fast-growing deep-tech company where your work has strategic impact and your decisions influence our product roadmap.
  • Enjoy autonomy, trust, and the opportunity to define the space-based AI datacenter ecosystem.

About the company

Magics is a European fabless semiconductor company that provides high-reliability, high-performance, fault-tolerant, and radiation-hardened chips for strategic sectors such as space, nuclear energy, aerospace, and defence markets. The company's products (Vision series, Motion series, AI series, and Time series) are designed for use in Nuclear applications and mission-critical systems, including robotics, sensors, 5G/6G communication satellites, drones, artificial intelligence and edge computing.

Apply for this position