Software Engineer, AI Tooling
MatXJob Title
Software Engineer, AI Tooling
Role Summary
Join MatX's AI Tooling team to build infrastructure and workflows that enable coding agents to operate safely and effectively across the company's hardware/software stack (ISA, RTL, simulator, compiler, kernels). The role focuses on agent identity, sandboxed execution environments, and a searchable agent knowledge base used by engineering teams.
Experience Level
Mid-level. No specific years of experience listed.
Responsibilities
Design and implement tooling and infrastructure that allow coding agents to act as first-class, auditable users of engineering systems.
- Provide agent identities, permissions, and audit trails for agent actions.
- Build and manage ephemeral and sandboxed execution environments (short-lived VMs, no-egress sandboxes).
- Aggregate and structure context from the monorepo, design documents, and Slack into a retrievable agent knowledge base.
- Integrate verification workflows so agents can run builds and simulations and report results securely.
Requirements
Core qualifications and practical skills required to perform the role.
- Production experience operating dynamic infrastructure and making maintainable architecture decisions.
- Extensive hands-on use of coding/LLM agents with clear policies for permitted actions and verification.
- Experience building engineering workflows, automation, review tooling, or agent harnesses adopted by other engineers.
- Ability to detect incorrect LLM outputs and update harnesses or verification to prevent recurrence.
Nice-to-have:
- Experience with Bazel remote execution or remote caching.
- Familiarity with sandboxing, ephemeral compute, or running untrusted workloads.
Education Requirements
Not specified.
About the Company
Company: MatX
Headquarters: Mountain View, California, USA
MatX specializes in creating faster chips for large language models (LLMs), focusing on innovative hardware and software solutions. The company fosters a collaborative and supportive work environment, welcoming candidates of all experience levels. Their approach prioritizes deep understanding and consideration of novel methods to drive efficiency and performance in their projects, particularly in silicon design and related engineering roles.
