UserRL trains agents with simulated users
In a post published on October 8, 2025, Salesforce presents UserRL, a framework for training agents with standardized environments and simulated users.
Source: UserRL treina agentes centrados no usuário com usuários simulados (x.com). Text prepared with AI from this source.
What happened and what to do
In a post dated October 8, 2025, Salesforce presents UserRL, a framework that combines standardized gym-style environments and simulated users to train agentic models. The post reports findings on cold starts with SFT, trajectory-based rewards, and scalable training with simulated users; these choices may inform agents designed for multi-step interactions.
A company could apply this approach by building test environments that reproduce real conversations and tasks with simulated user profiles. An implementation could measure task completion, response quality, and failures at each step, compare training strategies, and route critical cases for human review before expanding agent use.
How the consultancy can help
Wendelmaques can diagnose the relevant processes and risks, define metrics and scope for an evaluation environment with simulated users, and implement suitable testing, monitoring, and operations components on the company’s infrastructure.
Next step
Send a short description of the agent or process you want to evaluate to receive a scoped proposal.
Let's assess your project
I reply with scope and a proposal.