Performance Modeling Architect – AI Systems

Role Overview

We are looking for a Performance Modeling Architect who will develop the modeling and simulation frameworks used to guide architectural decisions across our platform. In thisrole, you will build models that capture the interaction between AI workloads, softwarestacks, and hardware architectures, enabling rigorous evaluation of system-leveltradeoffs. Your models will represent not only the SoC architecture and its components,but also key elements of the software stack and system behavior that ultimatelydetermine real-world performance.

These frameworks will enable hardware–softwareco-design, allowing us to explore design alternatives and quantify their impact on performance, latency, scalability, and energy efficiency. Your work will directly influence the architecture of next-generation AI compute platforms.

• Develop system-level performance models for AI workloads and compute architectures.
• Build simulation frameworks that capture interactions between AI models, software runtimes, and hardware systems.
• Model key components of SoC architecture including compute units, memory hierarchies, and interconnects.
• Incorporate relevant software and runtime behavior into modeling frameworks to reflect realistic system execution.
• Analyze system bottlenecks and evaluate architectural tradeoffs across the platform.
• Work closely with hardware architects to guide microarchitectural and system-level design decisions.
• Collaborate with software and ML teams to incorporate realistic workloads into modeling frameworks.
• Use modeling and simulation to evaluate performance, latency, throughput, and energy efficiency across target applications.

• Strong background in computer architecture, system architecture, or performance modeling.
• Experience building simulation or analytical models of complex hardware systems.
• Strong programming skills (Python, C++, or similar) for modeling and simulation frameworks.
• Understanding of modern compute architectures such as CPUs, GPUs, or AI accelerators.
• Ability to analyze complex systems and identify performance bottlenecks.
• Strong quantitative reasoning and ability to translate models into architectural insights.

• Experience modeling AI workloads or machine learning systems.
• Familiarity with architectural simulators or performance modeling tools.
• Experience with memory systems, interconnect architectures, or accelerator design.
• Exposure to hardware–software co-design methodologies.
• Experience modeling performance across multiple layers of the stack, including software runtimes and hardware platforms.

$200,000 - $500,000 a year

At Velaura, we believe exceptional talent deserves exceptional rewards. Compensation for this role includes a competitive base salary, performance-based incentives, andequity participation, allowing team members to share in the company’s long-term success.

Your base pay will depend on your skills, qualifications, experience, and location.

In addition to compensation, Velaura offers a comprehensive benefits package that may include medical, dental, and vision coverage; paid time off; flexible work arrangements; professional development opportunities; and other benefits designed to support the well-­being and growth of our team.

Velaura is committed to pay equity and transparency and regularly benchmarks compensation to ensure we remain competitive in the market.

Velaura is building next-generation compute technology for cloud, edge, and Physical AI. Our solutions will enable robots, autonomous systems, drones, and other intelligent machines to operate efficiently in the physical world. This is an opportunity to help build foundational technology at a time when the industry is undergoing fundamental change. You will work alongside experienced leaders, architects, engineers, and operators who have delivered industry-defining products across mobile, cloud, semiconductor, and AI platforms.

If you enjoy solving difficult problems, working across disciplines, and helping shape thefuture of Physical AI, we would love to hear from you.

Vela…