Architecture Modeling Engineer - AI Accelerator Performance

Arago; Paris; Sur place; EUR 90 000 - 130 000; Plein temps

Avantages offerts par ce poste

Stock options

Health benefits

Pension contributions

25 days PTO

Résumé du poste

Arago in Paris is seeking an Architecture Modeling Engineer to own performance modeling and workload analysis for its optical AI accelerator, collaborating across hardware and software to evaluate design choices and optimize kernels. You will translate workload requirements into architecture decisions and high-performance software on silicon.

The role requires strong math, computer architecture, and C/C++/Python skills, with experience in AI workloads and cross-team collaboration.

Qualifications

• Strong mathematical background in linear algebra, probability, numerical methods, and performance analysis.
• Deep knowledge of computer architecture, including pipelines, memory hierarchies, interconnects, and parallel execution.
• Proficiency in C/C++ and Python programming.
• Experience analyzing and optimizing performance-critical kernels or low-level workloads.
• Solid understanding of AI/ML workloads, tensor operations, data movement, and execution patterns.

Responsabilités

• Develop fast, flexible performance models to evaluate AI accelerator architecture choices.
• Characterize AI workloads, kernels, dataflows, and system bottlenecks with the software team.
• Model compute, memory, interconnect, and scheduling behavior across architectures with hardware teams.
• Conduct rapid performance studies and trade-off analyses across throughput, latency, utilization, bandwidth, and efficiency.
• Define and validate the programming model and workload mapping strategy for the accelerator.
• Collaborate with software team to optimize compute kernels on target hardware.
• Build lightweight simulation, profiling, and analysis tools to support fast iteration from architecture exploration to software optimization.

Connaissances

Math background

Computer architecture

C/C++ programming

Python programming

Performance analysis

AI/ML workloads

Cross-functional collaboration

English proficiency

Outils

Profiling tools

Description du poste

Arago in Paris is seeking an Architecture Modeling Engineer to own performance modeling and workload analysis for its optical AI accelerator, collaborating across hardware and software to evaluate design choices and optimize kernels. You will translate workload requirements into architecture decisions and high-performance software on silicon.

The role requires strong math, computer architecture, and C/C++/Python skills, with experience in AI workloads and cross-team collaboration.