Lead AI Processor Architect for Next-Gen Inference ASIC

European Tech Recruit; San Francisco (CA); Sur place; USD 250 000 - 420 000; Plein temps

Résumé du poste

European Tech Recruit is seeking a Senior Principal ASIC Architect to drive the architectural planning of next-generation AI processors. You will define chip-to-chip networking, including HBM and PCIe, and map AI workloads to hardware through detailed modelling and RTL verification.

You will mentor teams, coordinate with external partners, and establish major ASIC milestones to ensure peak performance, power efficiency, and manufacturability for hyperscale data center deployments.

Qualifications

• 15+ years hands-on silicon design engineering experience.
• Deep expertise in HBM and high-speed SerDes networking.
• Strong understanding of generative AI algorithms and transformer models.

Responsabilités

• Lead architectural planning for next-gen AI processors.
• Define chip-to-chip networking fabrics including HBM and PCIe.
• Analyse requirements and propose silicon solutions.
• Develop detailed design specs and lead modelling activities.
• Establish ASIC milestones ensuring functionality and efficiency.
• Collaborate with external partners and cross-functional teams.
• Provide technical guidance and mentor engineering teams.
• Evaluate silicon IP, EDA tools and design methodologies.

Connaissances

Silicon design

Architectural planning

RTL verification

Performance modelling

SystemC/TLM

Python scripting

C/C++

Verilog

EDA tools

HBM/SerDes

Outils

Verilog

RTL Design

SystemC

TLM

Python

Perl

Description du poste

European Tech Recruit is seeking a Senior Principal ASIC Architect to drive the architectural planning of next-generation AI processors. You will define chip-to-chip networking, including HBM and PCIe, and map AI workloads to hardware through detailed modelling and RTL verification.

You will mentor teams, coordinate with external partners, and establish major ASIC milestones to ensure peak performance, power efficiency, and manufacturability for hyperscale data center deployments.