Our Friendly client is reimagining the future of high-performance computing. Their accelerated compute solutions leverage intelligent adaptive algorithms to vastly accelerate large-scale systems and push supercomputers into a new generation.
With an innovative software-defined hardware architecture, they enable HPC & AI workloads to reach performance levels previously out of reach—and drive breakthroughs across advanced research fields.
The AI Compiler team plays a central role in enabling cutting-edge AI models and workflows to achieve world-class performance on their unique hardware architecture.
They are now looking for a skilled
AI Compiler Engineer
to help form a new team in Belgrade and contribute to building the compiler stack for next-generation platforms.
What You’ll Be Doing
-
Developing a next-generation MLIR-based AI compiler stack for specialized hardware
-
Working with platform teams to define IRs that leverage unique architectural properties
-
Collaborating with runtime teams on the evolution of the runtime stack, from drivers to integration with popular AI frameworks
-
Partnering with toolchain teams to deliver a cohesive and efficient toolchain supporting static and dynamic workflows
What We’re Looking For
-
4+ years of professional experience in software engineering
-
2+ years of experience building backend components for a large-scale, open-source-based compiler (e.g., LLVM)
-
At least one of the following:
-
Demonstrable open-source contributions to relevant projects
-
Experience developing state-of-the-art neural network architectures
-
Specialization in ML accelerators, HPC, GPU/CPU architectures
-
Experience with MLIR is a strong advantage
What You Get
-
Hybrid work model: 3 days in office + 2 remote
-
24 days of paid time off + public holidays
-
Private health insurance
-
Parking spot provided
-
Collaboration with a global team in a time zone aligned with Serbia
-
Stock options for all employees
If this sounds like the right next step for you, hit
Apply
and start the conversation.