Architecture Modeling Engineer

About the role

Arago is looking for an Architecture Modeling Engineer to lead performance modelling and workload analysis for its optical AI accelerator. You will work closely with both hardware and software teams to translate AI workload requirements into architectural decisions and high‑performance software on silicon.

Key responsibilities

• Develop fast, flexible performance models to evaluate accelerator architecture choices and guide early design decisions.
• Characterize key AI workloads, kernels, dataflows and system bottlenecks in collaboration with the software team.
• Model compute, memory, interconnect and scheduling behavior together with hardware architects.
• Conduct rapid performance studies and trade‑off analyses across throughput, latency, utilization, bandwidth and efficiency.
• Help define and validate the programming model and workload mapping strategy for the accelerator.
• Work hands‑on with software engineers to optimise compute kernels and key workloads for the target hardware.
• Build lightweight simulation, profiling and analysis tools to support fast iteration.

Required profile

• Strong mathematical background in linear algebra, probability, numerical methods and quantitative performance analysis.
• Deep knowledge of computer architecture, including compute pipelines, memory hierarchies, interconnects and parallel execution.
• Proficient in C/C++ and Python programming.
• Experience analysing and optimising performance‑critical kernels or low‑level compute workloads.
• Good understanding of AI/ML workloads, tensor operations, data movement and common execution patterns.
• Ability to collaborate across hardware and software teams and translate workload behaviour into architectural requirements.
• Professional proficiency in English.

Required skills

• C
• C++
• Python
• Linear algebra
• Probability
• Numerical methods
• Performance analysis
• Computer architecture
• AI/ML workload profiling
• Tensor operations
• Data movement optimisation
• Parallel execution
• Simulation and profiling tools