Senior Architecture & Performance Modeling Engineer (AI Inference Silicon)

Location: Remote (United States). Candidates must already be authorized to work in the U.S.; new visa sponsorship is unavailable, although H-1B transfers may be facilitated for eligible candidates.

Base salary: $220,000–$270,000 per year, excluding equity and other benefits.

Company

hirify.global develops AI inference accelerators and systems designed for low-latency, high-throughput large language model inference, including the Atlas, Asimov, and Titan platforms.

What you will do

• Partner with chip architects to define modeling studies, compare architectural alternatives, quantify tradeoffs, and recommend changes.
• Analyze compute, memory, dataflow, DMA, command processing, on-chip interconnect, and chip-to-chip communication.
• Evaluate prefill and decode workloads across batch sizes, context lengths, dense models, and mixture-of-experts models.
• Develop analytical, transaction-level, and cycle-level models with an appropriate balance of fidelity and iteration speed.
• Maintain reusable models in the native Rust simulator and correlate results with RTL simulation, emulation, and silicon measurements.
• Communicate findings, investigate unexpected results, and document assumptions, limitations, discrepancies, and regression coverage.

Requirements

• 5+ years of experience in computer architecture, microarchitecture, or performance modeling for accelerators, CPUs, or GPUs.
• Experience using quantitative models to identify bottlenecks, compare design alternatives, and influence architectural decisions.
• Strong understanding of pipelines, queues, parallel execution, memory hierarchies, data movement, and interconnects.
• Hands-on experience building analytical, transaction-level, or cycle-level performance models, including first-order estimates.
• Ability to work with incomplete specifications, make assumptions explicit, and balance modeling fidelity with iteration speed.
• Strong Rust skills, or expert C/C++ skills with willingness to work in Rust daily, plus Python for workload analysis, experiments, and visualization.
• Ability to read SystemVerilog RTL and verification results, reconcile them with architectural specifications, and calibrate models.
• Excellent communication skills for explaining results, confidence levels, and recommended next investigations or changes.

Nice to have

• Experience with LLM inference, including attention, KV caches, quantization, mixture-of-experts workloads, and multi-chip parallelism.
• Experience connecting workload behavior and software execution choices to hardware performance.

Culture & Benefits

• Remote-first work with a company-provided computer and home office setup.
• Company-paid medical, dental, vision, life, and disability coverage, including dependent coverage for medical, dental, and vision plans.
• Unlimited paid time off and 13 paid company holidays.
• 401(k) with company matching from day one, plus equity and other compensation components.