Friendly client is reimagining the future of high-performance computing. Their accelerated compute solutions rely on intelligent adaptive algorithms to dramatically speed up large-scale systems and unlock a new generation of breakthroughs in advanced research.

With a unique software-defined hardware architecture, they enable HPC & AI workloads to achieve performance previously considered out of reach.

They are now looking for an experienced expert who will build and lead this team from the ground up.

About the role

As the AI Compiler Lead, you will own the development of a state-of-the-art MLIR-based AI compiler stack targeting next-generation hardware.
This is a high-visibility, hands-on leadership role where you will assemble and mentor a local engineering team, while collaborating closely with global platform, runtime, and toolchain groups.

You will shape how cutting-edge AI models run on custom architectures—delivering breakthrough performance across HPC and AI workloads.

What you’ll be doing

• Develop a next-generation MLIR-based AI compiler stack targeting next generation hardware and delivering superior acceleration to AI applications
• Designing and defining IRs in collaboration with platform teams
• Partnering with runtime teams on the evolution of the runtime stack—from driver level to deep integration with major AI frameworks
• Working with toolchain teams to deliver a holistic experience for customers, supporting both static and dynamic development workflows
• Leading and mentoring a team of software engineers in Belgrade

What we’re looking for

• 6+ years of experience in software engineering
• 3+ years of experience contributing to or developing backend components for a large-scale, open-source-based compiler (e.g., LLVM)
• 2+ years leading a team of 3+ full-time software engineers
• At least one of the following:
• Open-source contributions to relevant projects
• Experience developing cutting-edge neural network architectures
• Expertise in ML accelerators, HPC, GPU/CPU architectures
• Backend experience with MLIR is a strong plus

What you get

• Hybrid work model: 3 days from the office + 2 days remote
• 24 days of paid time off + public holidays
• Private health insurance
• Parking spot provided
• Working with a global team in a time zone aligned with Serbia
• Stock options for all employees

If you want to lead a high-impact engineering team and build technology that sets new standards in HPC and AI, this is your place.