Compiler Engineer

Responsibilities

• Design and implement compiler passes across the Tungsten toolchain, including mid-end optimization, backend code generation, instruction scheduling, and register allocation.
• Co-design language constructs to improve expressiveness and performance for dataflow execution on wafer-scale hardware.
• Develop and iterate on code generation strategies for complex scientific and AI workloads to close the gap between peak and achieved throughput.
• Extend the compiler to support future hardware architectures as they transition from design to silicon.
• Collaborate directly with ASIC architects and application researchers to inform hardware-software co-design decisions.

Requirements

• Substantial experience in compiler development, specifically in IR design, optimization passes, or backend implementation.
• Strong grasp of computer architecture, including instruction sets, memory models, and dataflow execution.
• Proficiency in systems-level programming in C, with the ability to reason about performance at the instruction and memory-access level.
• Ability to approach compilation as a design problem, considering the interaction between language semantics, IR, and hardware capabilities.
• Excellent communication skills for working in a small, fast-moving, highly collaborative team.

Preferred Qualifications

• PhD in Computer Science or Computer Engineering.
• Experience with compilers for spatial, dataflow, or CGRA architectures.
• Exposure to ML compiler frameworks such as MLIR, XLA, or TVM.
• Experience with multi-dimensional data representations, tiling strategies, and vectorized operations.
• Experience building runtime systems, debuggers, or architecture simulators for non-standard hardware.

About the Company

Cerebras Systems builds the world's largest AI chip, 56 times larger than GPUs. Our novel wafer-scale architecture provides the AI compute power of dozens of GPUs on a single chip, delivering industry-leading training and inference speeds. Our Advanced Technology Group (ATG) is our pathfinding organization, exploring new architectures and shaping the technical roadmap for the future of wafer-scale systems.