AI Compiler Engineer – Lead Team in Belgrade

About the role

We are building a new AI Compiler team in Belgrade to create a cutting‑edge compiler stack that unlocks world‑record performance on our proprietary hardware. As the lead AI Compiler Engineer, you will design and implement the compiler infrastructure for next‑generation AI platforms, working closely with hardware, runtime, and toolchain groups.

Key responsibilities

• Develop a state‑of‑the‑art MLIR‑based AI compiler stack targeting our next‑gen hardware.
• Define intermediate representations (IRs) in collaboration with platform teams to exploit unique architectural features.
• Partner with runtime engineers to integrate the compiler with drivers and popular AI frameworks.
• Work with toolchain teams to deliver a complete development workflow for customers, supporting both static and dynamic compilation.

Required profile

• 6+ years of professional software engineering experience.
• 3+ years developing back‑ends for large‑scale, open‑source compilers such as LLVM.
• 2+ years leading a software group of at least three engineers.
• Demonstrated open‑source contributions, or experience with advanced neural‑network architectures, or specialization in ML accelerators, HPC, GPU or CPU architectures.

Required skills

• LLVM
• MLIR
• GPU architectures
• CPU architectures
• High‑performance computing (HPC)
• Neural‑network architecture development

What we offer

• High‑visibility, hands‑on role shaping the future of AI compilation.
• Opportunity to build and lead a new team from the ground up.
• Collaboration with world‑class hardware and software experts.