ASIC Architect, Principal (AI Inference)

Role Overview

We're looking for an exceptional Principal ASIC Architect to help define the future of AI inference hardware. You will collaborate across architecture, software, machine learning, systems, and silicon implementation to design next-generation AI accelerators capable of serving rapidly evolving frontier models. This role requires exceptional technical breadth, balancing mathematical understanding, ML trends, systems architecture, and practical silicon implementation. This may be the single most important technical hire after the Chief Architect.

Key Responsibilities

Architectural Exploration

• Own architectural exploration across the evolving landscape of transformer models, MoE, sparse models, and long-context inference.
• Evaluate emerging techniques including speculative decoding, continuous batching, KV-cache evolution, sub-quadratic and linear attention, and state-space models.
• Assess memory architectures, communication architectures, and interconnects for next-generation accelerators.

Performance Modeling

• Develop analytical models and build simulation frameworks to evaluate architectural tradeoffs.
• Estimate latency, throughput, bandwidth, memory and compute utilization, scaling efficiency, power, area, and cost.

Hardware/Software Co-Design

• Partner closely with compiler, runtime, system software, firmware, ASIC engineering, and ML research teams to align hardware direction with real workload needs.

ASIC Specification & Roadmap

• Develop architecture documents and define subsystem interfaces.
• Evaluate tradeoffs across competing architectural directions and guide the future product roadmap.

Industry Awareness

• Track new LLMs, reasoning models, research papers, hardware competitors, academic research, and emerging algorithms to keep Positron's architecture ahead of the field.

AI-Enabled Engineering

• Use AI to accelerate research, performance analysis, architecture exploration, simulation generation, and documentation.

Required Qualifications

• 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.
• Excellent written and verbal communication skills.

Preferred Qualifications

• Experience with GPU architecture.
• A solid grasp of ML systems, distributed inference, and large-scale serving.