Positron
ASIC Architect, Principal (AI Inference)
$300K–$350K
Remote (United States) · Remote · Principal · Full-time · Posted Jul 22, 2026
- Experience
- 15+ years in ASIC architecture.
- Visa sponsorship
- Must be currently authorized to work in the U.S.; no new visa sponsorship, but H-1B transfers may be facilitated for eligible candidates.
About the job
Lead architectural exploration and performance modeling for next-generation AI inference accelerators, shaping ASIC design and product direction in collaboration with software, ML, systems, and silicon teams.
What you'll do
- Evaluate model and inference techniques, memory architectures, communication architectures, and interconnects for future accelerators.
- Build analytical performance models and simulation frameworks to assess latency, throughput, utilization, scaling, power, area, and cost.
- Define ASIC architecture documents and subsystem interfaces, align hardware with software and workload needs, and guide the product roadmap.
What they ask for
- 15+ years of experience in ASIC architecture.
- Deep expertise in AI accelerator design and high-performance SoCs.
- Hands-on experience building performance models for memory systems and interconnects.
- Strong hardware/software co-design ability.
Nice to have
- Experience with GPU architecture.
- Grasp of ML systems, distributed inference, and large-scale serving; transformer internals; compiler experience; or research publications.
Pay
$300,000 – $350,000 (base salary)
Tools
ASIC, GPU, transformer, MoE, KV-cache, state-space models
AI focus
Architect AI inference accelerators for evolving models and inference methods; use AI to accelerate research, performance analysis, architecture exploration, simulation generation, and documentation.
Model-written summary · the original posting is authoritative