About the role
structured by ORIOur mission is to automate knowledge work. Sim is a platform to build autonomous AI agents for enterprise systems.
What you will do
- Own systems for context retrieval, planning and execution, tool use, memory, and recovery from failure.
- Build evals and use production traces to improve how the harness translates company knowledge into reliable agent behavior.
- Use findings to shape the harness architecture and the capabilities built next.
What they are looking for
- You can reason about how agents retrieve and use company context, including knowledge graphs, tool calls, memory, and context management, and translate that understanding into reliable systems.
- You can design evals, diagnose failures across retrieval, model behavior, and execution, and use evidence to improve task quality, latency, and cost.
Before you apply
- In-person role in San Francisco
- Hiring process includes technicals and an onsite small project in San Francisco office
- Visas are sponsored
Full posting text
Our mission is to automate knowledge work.
Sim is a platform to build autonomous AI agents for enterprise systems. Teams connect their integrations and Sim builds a knowledge graph of company data and processes; from there, they work with Sim to build and monitor the AI agents and workflows that run production systems.
Sim is talent-dense. We're looking for special people who are mission-driven, passionate, and creative.
We're a team of 10 working full-time, in person in San Francisco. We have a flat organization: you'll work directly with the founders and own your work from idea to execution. We sponsor visas.
ABOUT THE ROLE
The agent harness is Sim. It's the agent that ingests and reasons over a knowledge graph of company context to build and operate agents. You'll own the systems that make this possible: context retrieval, planning and execution, tool use, memory, and recovery from failure. You'll build evals and use production traces to improve how the harness translates company knowledge into reliable agent behavior. Those findings will shape the harness architecture and the capabilities we build next.
YOU MAY BE A FIT IF
- You can reason about how agents retrieve and use company context, including knowledge graphs, tool calls, memory, and context management, and translate that understanding into reliable systems.
- You can design evals, diagnose failures across retrieval, model behavior, and execution, and use evidence to improve task quality, latency, and cost.
New graduates are welcome. We care about what you can do; projects, internships, and independent work count.
COMPENSATION
$190K–$250K base plus a 0.25%–0.4% equity grant.
APPLYING
If there appears to be a fit, we'll reach out to schedule 1-2 short technicals. After, we'll schedule an onsite in our office, where you'll work on a small project, discuss ideas, and meet the team.
EQUAL OPPORTUNITY
Sim is committed to creating an inclusive environment for all employees. We celebrate diversity and are committed to creating an inclusive environment for all team members. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.