AI Accelerator Architecture Modeling Engineer

Arago; Paris; Sur place; EUR 85 000 - 120 000; Plein temps

Avantages offerts par ce poste

Stock options

Healthcare coverage

Pension contributions

Professional development

25 days PTO

Résumé du poste

Arago in Paris seeks an Architecture Modeling Engineer to own performance modeling for its optical AI accelerator. You will work across hardware and software teams to evaluate design choices, shape the programming model, and optimize kernels.

This role turns workload requirements into architecture decisions and high-performance software on silicon. The position emphasizes hands-on optimization, rapid prototyping, and close collaboration with hardware and software teams to accelerate AI workloads

Qualifications

• Strong mathematical background with numerical methods for performance analysis.
• Solid knowledge of computer architecture including pipelines, memory, interconnects, and parallel execution.
• Proficiency in C/C++ and Python programming.
• Experience optimizing performance-critical kernels or low-level compute workloads.
• Familiarity with AI/ML workloads, tensor operations, and data movement patterns.
• Ability to translate workload behavior across hardware and software teams into requirements.

Responsabilités

• Develop fast performance models to evaluate AI accelerator architecture choices.
• Characterize AI workloads, kernels, dataflows, and bottlenecks with the software team.
• Model compute, memory, interconnect, and scheduling behavior across proposed architectures with hardware architects.
• Conduct rapid performance studies and trade-off analyses (throughput, latency, utilization, bandwidth).
• Help define and validate programming model and workload mapping for the accelerator.
• Collaborate hands-on with software to optimize kernels and workloads for target hardware.
• Build lightweight simulation, profiling, and analysis tools to speed architecture exploration.

Connaissances

Math background

Computer architecture

C/C++ & Python

Performance analysis

AI/ML workloads

Cross-team collaboration

English proficiency

Description du poste

Arago in Paris seeks an Architecture Modeling Engineer to own performance modeling for its optical AI accelerator. You will work across hardware and software teams to evaluate design choices, shape the programming model, and optimize kernels.

This role turns workload requirements into architecture decisions and high-performance software on silicon. The position emphasizes hands-on optimization, rapid prototyping, and close collaboration with hardware and software teams to accelerate AI workloads