Software Engineer - AI Framework

Software Engineer – AI Framework
Hybrid Work | Full-Time

An innovative semiconductor team is seeking a Software Engineer with deep expertise in AI and machine learning frameworks to help build the next generation of GPU and AI acceleration solutions. This is a unique opportunity to contribute to cutting-edge technology focused on delivering high-performance and energy-efficient compute platforms for modern AI workloads.

You'll be working on a flagship GPU and AI platform supporting PyTorch, OpenCL, and Vulkan, designed to bring scalable, efficient AI capabilities to developers and researchers across the industry.

Role Overview:
As a Software Engineer – AI Framework, you’ll be responsible for developing and optimizing PyTorch integrations and other AI tools for a custom AI accelerator platform. You’ll work closely with hardware and software teams to ensure tight integration, performance tuning, and a seamless developer experience.

Key Responsibilities:

• Develop and maintain PyTorch integration for a custom AI platform.
• Build and optimize kernels and libraries for core AI operations.
• Create tooling for neural network deployment and optimization.
• Benchmark, analyze, and improve AI workload performance.
• Collaborate with the hardware team to guide architectural decisions.
• Extend support to additional frameworks (e.g., TensorFlow, ONNX).
• Produce developer documentation and resources.

Requirements:

• 5+ years of experience in AI/ML software development.
• Deep understanding of PyTorch internals and other major ML frameworks.
• Experience optimizing deep learning performance on accelerator hardware.
• Solid knowledge of deep learning algorithms and compute patterns.
• Strong programming skills in C++, CUDA, or OpenCL.
• Background in performance profiling and optimization.
• BS/MS in Computer Science, Electrical Engineering, or a related field.

Interested?
Send your CV to Trei@microtech-global.com to apply.