About the role
Sepal AI is working with top-tier AI research labs (OpenAI, Anthropic, Google) to ensure next-generation AI systems behave safely and reliably inside real-world enterprise software. We’re building safety benchmarks that simulate business workflows—like Salesforce—and uncovering how AI models might help or fail in high-stakes environments.
🧠 What You'll Do
- Recreate Salesforce workflows (e.g., Lead-to-Cash, CPQ) to test AI-assistant behavior.
- Design realistic tasks and edge cases where AI could make risky or costly errors.
- Create structured rubrics to evaluate whether AI decisions are safe, compliant, and reliable.
- Collaborate with AI researchers to refine safety tests and improve model alignment with enterprise needs.
✅ Who You Are
- 3+ years configuring or analyzing Salesforce systems.
- Strong with declarative Salesforce tools (Flows, Validation Rules, Reports).
- Familiar with risk-prone workflows like quoting, deal desk, or approvals.
- Bonus: background in QA, process documentation, or internal audits.
💸 Pay: $30–60/hr
🌍 Remote
⏱ Flexible hours
🧭 Impact: Your expertise defines the standards AI must meet before it can be trusted in critical business workflows.
Education: Bachelor's degree completed
Official source: expert-hub.sepalai.com. Details can change; the platform's page is the final word. Something wrong with this listing? Tell us.
More coding jobsMore AI agent and RL task jobs All Sepal AI roles How much does Sepal AI pay?
