Remote AI Inference Simulation Engineer

Positron; United States; Remote; USD 200,000 - 300,000; Full time

Benefits offered by this job

Unlimited PTO

Remote first culture

401(k) with company matching

Company-paid health insurance

Job summary

Positron AI is seeking a qualified engineer to build simulation and modeling infrastructure for AI inference systems. You will develop analytical performance models, timing simulators, and virtual platforms to explore architectures and support firmware, runtime, and host software workloads.

The role combines computer architecture and software engineering, emphasizing fidelity, determinism, and speed, with the opportunity to influence hardware decisions through reproducible experiments.

Qualifications

• Strong software engineering in a systems language (C++/Rust).
• Deep understanding of processors, memory hierarchies, and interconnects.
• Experience designing/validating models with fidelity, determinism and speed.
• Mathematics and quantitative reasoning to turn workloads into models and experiments.
• Ability to integrate complex software systems and debug across components.
• Clear technical writing and documenting assumptions and tradeoffs.
• BS, MS, or PhD in CS, EE, or related field.

Responsibilities

• Develop analytical performance models and timing simulators.
• Build functional and timing models of accelerator compute and memory.
• Integrate accelerator models with virtual platforms and host interfaces.
• Compare architectures with reproducible experiments and document assumptions.

Skills

C++/Rust

Systems thinking

Quantitative modeling

Debugging/verification

Software integration

Technical writing

BS/MS/PhD CS/EE

Education

BS, MS, or PhD in CS/EE

Tools

SystemC

Arm Fast Models (LISA)

QEMU

gem5

Perfetto

Job description

Positron AI is seeking a qualified engineer to build simulation and modeling infrastructure for AI inference systems. You will develop analytical performance models, timing simulators, and virtual platforms to explore architectures and support firmware, runtime, and host software workloads.

The role combines computer architecture and software engineering, emphasizing fidelity, determinism, and speed, with the opportunity to influence hardware decisions through reproducible experiments.