About the Role
As an Applied AI Engineer, you will build, refine, and scale agent systems within Sauna, focusing on architecture, evaluations, and deployment to ensure production-grade performance, reliability, and traceability.
Responsibilities
- Implement multi-step, tool-using agents that handle real APIs, retries, authentication, and edge cases.
- Design agent memory systems to persist state across runs, including memory migrations and context organization.
- Develop agents capable of proactive task execution and user notifications.
- Own and evolve the evaluation framework, including automated checks and human-in-the-loop scoring.
- Work across infrastructure, frontend, and product to ensure agent functionality.
Requirements
- 3+ years of engineering experience, including shipping production software.
- Experience building and deploying agent-like systems such as multi-step LLM pipelines or tool-using bots.
- Hands-on experience with agent orchestration, memory management, and tool use.
- Experience with evaluation metrics, regression testing, and improving agent behavior.
- Ability to write production-grade code and work across systems independently.
- US work authorization.
- Must be based in the Bay Area or willing to relocate before starting.
Skills
- Agent orchestration
- Memory management
- Tool use and API integration
- Evaluation frameworks
- LLM ops
- Tracing and observability
- Production software engineering
Location
- San Francisco, CA (Presidio)
Work Type
- Full-time
- In-person
- 5 days a week
Experience Level
- 3+ years
Salary/Compensations
- $180K–$240K
Benefits
- Meaningful early-stage equity
- Health insurance
- Dental insurance
- 401(k)
- Considerable PTO
- Gym budget
- Lunch
About the Company
- Wordware is a ~20-person team building Sauna, an AI coworker that learns how you work.
- Backed by Spark Capital, Felicis, and Y Combinator with a $30M seed funding.
