A company developing AI-focused processors, semiconductor IP, and software solutions is looking for a Compiler Software Engineer. This role develops and optimizes compiler technologies and software toolchains that enable machine learning models to run efficiently on proprietary AI accelerator hardware.
Responsibilities
Develop and enhance AI compiler frameworks and software toolchains.
Optimize machine learning graphs and model execution for proprietary AI accelerators.
Design and implement compiler features using C++ and Python.
Collaborate closely with hardware architects, machine learning engineers, and software teams to maximize AI performance and efficiency.
Support code generation, optimization, and deployment of machine learning models.
Debug and resolve complex compiler and runtime issues.
Contribute to CI/CD workflows, software quality initiatives, and innovation in AI inference optimization and Edge AI deployment.
Requirements
Required Qualifications
Bachelor's degree in Computer Engineering, Computer Science, Electrical Engineering, or a related field.
5+ years of software engineering experience.
3+ years of professional C++ development experience.
2+ years of Python development experience, including package development and C++ bindings.
2+ years of experience with machine learning graph optimization.
Experience using MLIR and LLVM frameworks.
Experience with Git-based development workflows and GitHub/GitLab environments.
Familiarity with CI/CD tools such as Jenkins.
Understanding of machine learning quantization and mixed-precision inference.
Strong debugging, analysis, and problem-solving skills.
Benefits
Full remote work available.
Comprehensive social insurance coverage.
Opportunity to work on cutting-edge AI processors and semiconductor technologies.
International and collaborative working environment.
Career growth opportunities within a fast-growing deep-tech organization.
Smoke-free workplace.
About the Company
A fabless semiconductor design company developing and providing AI-focused processor IP cores and software. Its mission is to address the fundamental mismatch between modern AI software and deep neural networks and conventional processors such as CPUs and GPUs, enabling more power-efficient, cost-effective, and faster AI processing at the edge with cloud-level performance.