Silicon Engineer

Anthropic•San Francisco, NY

2d•Hybrid

About The Position

Anthropic runs some of the largest AI training and inference workloads in the world, across multiple hardware platforms. We work from the chip level up with our silicon partners, and we are now deepening that investment by building a custom silicon team. You will be one of a small number of people covering a large surface. We are opening searches across the key domains below, and we are interested in candidates whose center of gravity sits in any of them: Front-end design Pre-silicon verification Physical design Design for test Analog and mixed-signal Technology and foundry CAD and design infrastructure Packaging, signal integrity, and power integrity This is a role for someone who has shipped silicon, has a realistic relationship with schedules, and is comfortable making consequential calls without a large organization behind them.

Requirements

Deep, hands-on expertise in at least one silicon domain: front-end design, pre-silicon verification, physical design, design for test, analog and mixed-signal, technology and foundry, design infrastructure, or packaging and signal/power integrity
Direct personal contribution to silicon that taped out and shipped, with ownership you can speak to in detail
Working fluency in the domains adjacent to your own, sufficient to reason about chip-level tradeoffs outside your specialty
Experience working with external partners — ASIC houses, IP vendors, foundries, or test partners — including reviewing their work and holding a technical bar
Track record of owning directional technical decisions and their consequences, not only making recommendations
Practical experience using AI coding tools in your own work, and clear ideas about where they help and where they do not
Clear written and verbal communication, and experience driving alignment across teams and organizations
Comfort operating with high autonomy and little scaffolding on a highly dynamic program

Nice To Haves

Experience on machine learning accelerators, high-performance compute, or other large, high-bandwidth designs
Experience on a founding or early-stage silicon team, including standing up flows, methodology, or tooling from nothing
Experience with both partner-executed and fully in-house programs, and a view on the tradeoffs between them
Familiarity with the chip-package-system interface and the tradeoffs that cross it
Experience with hardware-software co-design and working directly with compiler, kernel, or runtime teams
Experience building or extending AI-assisted design, verification, or physical design flows
Depth in memory subsystems, on-die interconnect, numerics, or low-power design

Responsibilities

Take part in end to end activities for the program: strategy, specification, execution and the technical bar.
Write the specifications and interface definitions that others design against and keep them coherent as the architecture converges
Make and own directional calls: technology and IP selection, where we build versus buy versus license, and where we take design authority versus rely on partners
Build and share AI-assisted approaches to the code-like parts of your domain, and help the team learn what Claude can genuinely accelerate
Contribute to the architecture and micro-architecture discussion from your domain's perspective, including the tradeoffs your area imposes on the rest of the chip
Partner with Anthropic's inference, performance, kernels, and infrastructure teams on hardware-software co-design, and with our hardware systems team on bring-up and integration
Support first-silicon bring-up and debug when it arrives, regardless of which domain you came in through