Staff AI Engineer, Agent Memory and Knowledge Graphs

On-site (Redwood City, CA) | $210K to $280K

PLEASE INCLUDE YOUR LINKEDIN TO BE CONSIDERED!

About the company

A well-funded, seed-stage enterprise AI company building an autonomous agent that investigates and heads off production incidents for SRE, platform, DevOps, and on-call teams. The product is already live with enterprise customers, working through alerts at real scale with strong root-cause accuracy, and the company is going after the broader observability market. The founders are on their third company, with two earlier startups acquired by large public tech companies, and much of the current team followed them from a previous venture. The seed round is unusually large and includes a major strategic tech investor.

The role

You'll be a senior individual contributor on the core agent, owning the hardest technical problems in how it understands a customer's environment. The center of the work is designing how the agent builds a graph of a customer's services and dependencies, and building the memory that lets it get sharper with every investigation it runs for that customer. You'll work side by side with the founders and the product manager, and you'll have a real vote on where the agent and its architecture go next. There are no direct reports. Your influence comes from design reviews, standards, and the code you ship.

What you'll do

• Design and build the algorithms that turn a customer's production environment into a knowledge graph or service map the agent can reason over during an incident

• Own the agent's memory layer so it retains what it learns about each customer's systems and investigates future incidents faster and more accurately

• Improve the context and evidence systems that tie the agent's root-cause conclusions to real signals across logs, metrics, traces, and change events

• Take ambiguous problems from framing and algorithm design all the way through production deployment, with evaluations gating what ships

• Shape the agent's product direction and next platform architecture alongside the founders and the PM

• Raise the bar for the AI engineers around you through design reviews, shared platform standards, and hands-on code

What you must bring

• 4+ years of experience as an AI or ML engineer

• A track record of building and deploying end-to-end ML or LLM systems that run in production

• Hands-on experience building agentic systems, such as software engineering agents or agent memory

• Classic ML models you've trained and shipped to production, not only LLM-based work

• The ability to move between research and engineering, and to carry research ideas into a working product

• A Master's degree in Computer Science

• Comfort operating with high autonomy as the senior technical owner of a problem area

• Ability to work on-site in Redwood City, CA five days a week

Nice to have

• Experience building knowledge graphs, service maps, or retrieval over operational data

• Exposure to observability, AIOps, or SRE tooling (logs, metrics, traces, incident workflows)

• Experience with agent orchestration and evaluation-driven development

• Production AI work at a frontier AI lab, a large tech company's AI org, or a high-growth startup

Compensation & logistics

• $210K to $280K base salary plus negotiable equity

• On-site five days a week in Redwood City, CA, with intense, startup-paced hours

• No relocation support and no travel

• Open to visa sponsorship, including new H-1B and TN, and to visa transfers, including OPT and H-1B transfers

This isn't the right fit if

• Your background is mainly research, without ML or LLM systems you've shipped to production

• Your AI experience is limited to LLM applications, with no classic ML models taken to production

• You want to lead through managing people rather than as a hands-on IC

• You need a remote or hybrid setup, or you're looking for steady, predictable hours