AI Software Development Engineer (Compiler)

TL;DR

AI Software Development Engineer (Compiler): Developing graph-based compilation algorithms and optimizing neural network performance for AI hardware accelerators with an accent on MLIR infrastructure and OpenVINO toolkit. Focus on designing scalable compilation flows, implementing optimization passes, and profiling performance for next-generation visual processing units.

Location: Must be based in Timisoara, Romania (Hybrid work model)

Salary: 171,550 RON – 409,550 RON

Company

A global leader in semiconductor technology and AI software solutions, driving innovation in computing and data center infrastructure.

What you will do

• Drive the design of compilation flows for future neural network architectures.
• Implement compilation passes to enable and optimize neural network workloads.
• Develop performance analysis, visualization, and profiling tools.
• Integrate and test changes, owning task execution end-to-end.
• Contribute to the planning of compilation features and roadmap.

Requirements

• BS/MS in Computer Science or a related field.
• Minimum 3-4 years of professional software engineering experience.
• Excellent C++ programming skills.
• Strong background in production software engineering, including CI, code reviews, and testing.
• Proven experience contributing to large-scale, multi-component software systems.
• Must be based in or able to work from Timisoara, Romania.

Nice to have

• Experience with compiler technologies (LLVM/MLIR).
• Knowledge of graph-based algorithms (graph coloring, shortest path, etc.).
• Experience with AI hardware accelerators, GPU, or heterogeneous architectures.
• Familiarity with deep-learning frameworks and mapping neural networks to hardware.
• Python programming proficiency.

Culture & Benefits

• Hybrid work model allowing flexibility between on-site and off-site work.
• Competitive total compensation package.
• Opportunity to work on cutting-edge AI hardware and software stacks.
• Collaborative environment working with research, runtime, and infrastructure teams.