Home › Computer Hardware Engineer jobs in the United States › Whirl AI
Staff Engineer, AI Platform
Whirl AI · Remote (US) · Posted November 3, 2026 (16 days ago)
- Remote (US)
- Full time
- $210,000–$280,000 a year
Job description
Company Description
Whirl AI helps Enterprise IT organizations deliver business change faster. Our platform gives enterprise teams AI agents grounded in a deep, continuously updated understanding of how their systems actually work - compressing discovery and documentation from weeks to hours without sacrificing quality or control. We are early-stage, moving fast, and building the team that will define how Enterprise IT operates in the AI era.
The Role You'll build the AI platform our agents run on, end to end: the data and context layer that lets them understand a customer's enterprise systems, and the harness that lets them call tools, work through multi-step tasks, take action safely and get measurably more accurate over time. Agents are only as good as the context they can trust and the harness around them, so you'll own both.
As a Staff Engineer, you'll be one of the technical leads for the platform, set the standards other engineers build on, and work directly with customers to turn messy enterprise IT requirements into systems that work.
This role is remote-eligible for anyone based in and authorized to work in the U.S. If you’re in the Bay Area, you’re also welcome to join us at our Financial District office in San Francisco.
Responsibilities
- Design and build the agent harness: tool calling, multi-step workflows, durable execution, and how agents recover when a step fails partway through.
- Build the data lake that lets agents find, understand and act on customer systems such as Salesforce and ServiceNow, including metadata, dependencies and the logic that actually runs the business.
- Contribute to the data platform underneath: data modeling, and keeping context current and accurate as those systems change.
- Define how we measure agent quality, with evals, tracing and accuracy against ground truth, and use those measurements to make the product better.
- Set guardrails and governance for agents working with sensitive enterprise data: access control, approvals, tenant isolation, audit ability and stopping conditions.
- Lead architecture and design reviews, write the technical documents others build from, and mentor senior engineers.
- Work with customers and Product to turn ambiguous requirements into a platform plan, and take part in technical conversations with key customers.
Required qualifications
- You've built and shipped agent systems to production: LLM agents that call tools, take actions in real systems and handle failure. You can walk us through one in detail.
- You can explain how you measured whether an agent system improved over time, and what you changed because of it.
- You've designed guardrails for agents that act on real data: permissions, human approval steps, termination conditions and cost controls.
- You've been the technical lead on a distributed system, data-intensive product or developer platform in production, and other engineers have built on your work.
- You've built for enterprise customers, where security, access control and governance shape the design.
- You have a well-founded view on API design, data modeling and system reliability, based on systems you've built and run.
- You use AI coding agents in your own work every day and can explain what you delegate and what you keep.
Preferred qualifications
- Experience with Salesforce, ServiceNow or other enterprise platforms, especially their APIs, metadata and automation layers.
- Open-source contributions to AI, agent, data or developer tooling projects.
What you'll work with
Python, Postgres, DuckDB, Pydantic AI data agents. LLM APIs, AWS, with infrastructure as code in Pulumi.
Benefits
- Own meaningful technical decisions at the ground floor of an AI company targeting a massive market
- Work directly with founders and have immediate, visible impact
- Competitive compensation including equity and benefits (100% employer-paid health plan coverage for employees)
Whirl AI is committed to equal employment opportunity. All hiring decisions are made on the basis of qualifications, merit, and business need - never on race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, veteran status, or any other characteristic protected by law.
We build better products because our team doesn't all think the same way, and we want that to keep being true. Show up as yourself. We can't wait to meet you.
Posting a65a2f25-caed-4db5-8aee-161a49be3975 on ashby. Source: the employer’s listing. Valid through December 18, 2026 unless closed earlier.