Internal Acceleration, Kernels

Job Title

Internal Acceleration, Kernels

Role Summary

The Internal Acceleration team builds agentic tools and operational systems to speed production of frontier tokens and internal engineering workflows. You will embed with kernel, ASIC, hardware, and production teams and work directly with founders to deliver internal products that materially increase engineering throughput.

This is a results-oriented role focused on shipping end-to-end internal tooling and agent integrations that engineers rely on daily.

Experience Level

Mid-level - 1–5+ years building and shipping software end-to-end, ideally at a fast-moving startup.

Responsibilities

Deliver agent-driven, full-stack internal tools from idea to production and measure their impact.

• Embed with kernels, ASIC, hardware, and production teams to identify high-leverage bottlenecks and own solutions end-to-end.
• Design, build, and deploy LLM agents integrated with engineering workflows via tool-calling and multi-component protocols.
• Ship full-stack internal products: UIs, data pipelines, and infrastructure for engineering productivity.
• Design evaluations, track adoption, and iterate on tools based on measured impact.
• Partner with founders on time-sensitive special projects and deliver mission-critical results.

Requirements

Core qualifications and relevant experience for immediate contribution.

• Must-have: 1–5+ years building and shipping software end-to-end; strong full-stack development skills.
• Must-have: Practical experience building LLM agents or AI tooling that real users depend on (context engineering, tool integration, evals, failure analysis).
• Must-have: Comfortable using agentic coding tools and frontier models to ramp into unfamiliar domains.
• Must-have: Product sense, design judgment, attention to detail, and strong ownership/agency.
• Nice-to-have: Exposure to chip development, kernels, hardware, or manufacturing.
• Nice-to-have: Infrastructure, data engineering, DevOps, fine-tuning, RAG over proprietary technical data, or multi-agent orchestration experience.
• Nice-to-have: Startup experience during hypergrowth.

Education Requirements

Not specified.

About the Company

Company: Etched

Headquarters: San Jose, CA, United States

Etched develops purpose-built AI inference ASICs and systems optimized for transformer models, aiming to deliver significantly higher performance, lower cost, and lower latency than GPUs. The company focuses on enabling applications like real-time video generation and advanced reasoning agents, and is backed by leading investors and engineers.