Senior/Principal Performance Modeling Engineer

Senior / Principal Software Engineer — Performance Modeling

Location: Santa Clara, CA

Employment Type: Full-time

Workplace: On-site

Team: Engineering

Leveling for this role will be based on scope, ownership, and technical leadership, not years of experience alone.

About the Company

We are a VC-backed, stealth-mode startup building rack-scale AI inference systems for the next generation of datacenter workloads.

Our differentiated SoC architecture enables system-level innovations that improve efficiency, throughput, and cost for large-scale inference serving. We are building both the hardware and the software stack required to serve leading-edge models with extreme efficiency.

This is an opportunity to join early and help shape a deeply technical company at the intersection of computer architecture, systems software, accelerators, and AI infrastructure.

About the Role

We are looking for a Senior or Principal Software Engineer to help build the simulation infrastructure that drives architecture decisions and accelerates hardware–software co-design.

You will design and develop functional and/or performance models for processors, memory systems, interconnects, and accelerator components. These models will be used to evaluate architectural tradeoffs, guide optimization, validate design assumptions, and improve the efficiency of the full system before hardware is available.

Depending on experience, you may lead a small team or serve as a senior individual contributor owning critical modeling and simulation infrastructure end to end.

This is a high-ownership role in a fast-moving startup environment. The work you do will directly influence architecture, hardware design, compiler strategy, and system-level performance.

What You’ll Do

• Design, implement, and maintain processor simulators, including performance and/or functional models
• Model key microarchitectural components such as pipelines, memory hierarchies, interconnects, and accelerators
• Generate, curate, and refine traces to support targeted performance analysis and architecture exploration
• Analyze simulation results to inform architecture tradeoffs, performance bottlenecks, and optimization opportunities
• Collaborate closely with architecture, hardware, compiler, and systems software teams
• Own critical simulation infrastructure from early concept through production use
• Help establish the modeling methodology, validation strategy, and engineering practices for a new hardware–software platform

You May Be a Fit If You Have

• 5+ years of software engineering experience, ideally in systems, computer architecture, performance modeling, or EDA
• Strong C/C++ and Python skills, with experience working in large and complex codebases
• Experience with processor simulators, architectural modeling, or performance analysis
• Hands-on experience developing simulation infrastructure from scratch or substantially extending existing simulators
• Solid understanding of computer architecture and microarchitecture
• Ability to work independently, make technical decisions, and drive ambiguous projects in an early-stage startup environment

Strong Candidates May Also Have

• Experience with AI/ML accelerators, GPUs, SoCs, or heterogeneous systems
• Familiarity with gem5, gem5-SALAM, proprietary modeling tools, or internal simulator frameworks
• Background in hardware–software co-design, compiler interactions, or architecture exploration
• Experience modeling memory systems, interconnects, pipelines, or accelerator datapaths
• Experience building infrastructure for trace generation, trace replay, workload characterization, or performance analysis
• Comfort working across architecture, RTL, compiler, firmware, and systems software boundaries

Why This Role Matters

In advanced AI systems, architecture decisions are only as good as the models that guide them.

Simulation is the bridge between ideas and silicon. The models, traces, and analysis infrastructure you build will help determine which architectural bets are worth making, where performance is won or lost, and how hardware and software should evolve together.

As an early engineer, you will have significant technical ownership, work alongside a highly technical team, and help shape the foundation for a new class of rack-scale AI inference systems.