AI/ML Engineer

Astera Labs•San Jose, CA

9d•Onsite

About The Position

Astera Labs is seeking an AI/ML Engineer to build production AI systems for technical users. This applied engineering role focuses on transforming modern model capabilities into reliable, usable systems. Key challenges include managing context, ensuring reliable tool use, designing effective abstractions for skills and workflows, developing relevant evaluations, and iterating to achieve daily workflow integration. Potential projects include coding agents for terminals and IDEs, verification and debug assistants, log-analysis systems for product diagnostics, documentation and spec-comparison agents, and internal assistants leveraging company knowledge and engineering data. The role requires an end-to-end perspective, encompassing prompt and context design, retrieval quality, tool interfaces, evaluations, failure modes, deployment, and continuous improvement.

Requirements

1–5 years of experience in software engineering, applied AI, ML engineering, or related backend/platform roles.
Strong Python skills and strong production engineering fundamentals.
Hands-on experience building AI/LLM applications, agents, retrieval-backed systems, or workflow automation.
Comfort working with tool-using systems where correctness depends on context quality, tool integration, and careful failure handling.
Experience with AWS or GCP and the realities of deploying and debugging production AI services.
Good judgment around evals, failure modes, latency/cost tradeoffs, and safe rollout of non-deterministic systems.
Clear communication and the ability to turn ambiguous technical workflows into robust product behavior.

Nice To Haves

Have built more than demos.
Have worked on systems where retrieval quality matters, where tool use can fail in subtle ways, where evaluation changes engineering decisions, and where product usefulness depends as much on system design as on model choice.
Care about the details that separate a clever prototype from a dependable system.

Responsibilities

Build AI applications and agentic workflows for engineering productivity, diagnostics, search, documentation, and workflow automation.
Design systems that combine LLMs with retrieval, tool use, structured outputs, and evaluation loops.
Integrate models with internal tools, APIs, CLIs, MCP interfaces, and operational workflows to perform tasks in real environments.
Improve system quality through eval design, prompt and context iteration, model selection, failure analysis, and human feedback.
Build reusable skills, workflows, and abstractions for sharing capabilities across agents and teams.
Work closely with infrastructure and domain teams to deploy, monitor, and continuously improve AI systems in production.