Related to an AI-coordinated HRIS:

For an AI layer mediating front-end requests and SQL data, design the agent to issue small, incremental SQL queries and adjust each next step based on returned results (“doing more”) instead of crafting one large query plan upfront (“thinking more”). The Test-Time Interaction (TTI) work by Mialon et al. shows that scaling interaction—short query–feedback loops—beats longer chain-of-thought reasoning under the same compute budget, letting agents refine reports, validations, or rewards programs on the fly and learn efficient patterns over time.

Reference
Mialon, G.; El Mhamdi, E.; et al. Thinking vs. Doing: Agents that Reason by Scaling Test-Time Interaction. 2024.