AI Compiler Engineer

The Role:

As an AI Compiler Engineer, you will be responsible for designing, developing, and optimizing the AI compiler toolchain that bridges high-level machine learning frameworks and Openchip’s AI hardware accelerators (RISC-V). You will work closely with software, hardware, and AI engineers to enable efficient inference execution of best-in-class Gen

AI LLMs, optimizing code generation and advance compiler infrastructure.

Your work will directly influence performance and scalability across Openchip’s AI inference stack.

Key Responsibilities:

· Design, implement, and maintain compiler components for AI inference workloads, focusing on graph lowering, optimization, and code generation (from frameworks like Py

Torch, Tensor

Flow, or ONNX to hardware).

· Collaborate with hardware teams to leverage hardware features, maximizing performance.

· Contribute to open-source compiler ecosystems (e.g., LLVM, MLIR, TVM, XLA, IREE).

· Profile, debug and optimize model execution to maximize throughput and minimize latency together with energy efficiency.

· Enable validation on prototype silicon or simulators or real hardware.

Required Qualifications:

· Master’s or Ph

D in Computer Science, Electrical/Computer Engineering, or related field.

· 3 years of experience in compiler development for AI/ML, GPU, DSP, or custom accelerators.

· Proven expertise with MLIR, LLVM, or similar compiler infrastructures.

· Strong programming skills in C and Python.

· Solid understanding of neural network computation graphs, linear algebra, and AI inference frameworks.

· Experience in code generation, optimization passes, and runtime integration.

Technical skills:

· MLIR / LLVM / IREE / TVM compiler development

· Model formats: ONNX, Tensor

Flow, Py

Torch

· Low-level optimization (SIMD, vectorization, scheduling, data-