Performance Architect

Performance Modeling / Architecture Simulation Engineer

Acceler8 Talent is partnering with a stealth-mode startup building next-generation AI inference systems optimized for data center-scale deployment. Their architecture combines differentiated silicon and system-level innovation to maximize efficiency for large-scale inference serving. We're seeking highly skilled engineers to help build hardware platforms and develop software infrastructure capable of serving leading-edge AI models with exceptional efficiency.

Lead or contribute to a small engineering team developing performance and/or functional models that guide architectural decisions and accelerate hardware/software co-design in a fast-moving startup environment.

• Design, implement, and maintain processor simulators, including performance and functional models
• Model microarchitectural components such as pipelines, memory hierarchies, interconnects, and accelerators
• Generate and curate traces suitable for targeted performance analysis and architectural studies
• Analyze simulation results to drive architecture trade-offs and optimization decisions
• Collaborate closely with architecture, hardware, and compiler teams
• Own critical simulation infrastructure from initial concept through production deployment

• 5+ years of software engineering experience, ideally within systems, computer architecture, or EDA-related domains
• Strong C/C++ and Python development skills working with large, complex codebases
• Experience with processor simulators, architectural modeling, or performance analysis, particularly involving from-scratch simulator development
• Strong understanding of computer architecture and microarchitecture
• Ability to work independently and make technical decisions in an early-stage startup environment

• Experience with AI/ML accelerators, GPUs, or heterogeneous compute systems
• Familiarity with simulators such as gem5, gem5-SALAM, or proprietary architectural modeling environments
• Background in hardware/software co-design or compiler technologies

• Opportunity to shape next-generation AI inference infrastructure from the ground up
• High-impact technical ownership in a fast-moving engineering environment
• Work at the intersection of AI systems, silicon architecture, and large-scale compute infrastructure