Resources · Glossary
The language ofagentic recruiting.
Plain-English definitions for the terms behind AI-native hiring: what they mean, and how they work in Talara.
Agentic ATS
An applicant tracking system where an AI agent does the recruiting work using many tools, while a human approves every consequential decision. Read the full definition.
AI-native ATS
An applicant tracking system built around AI from the ground up, rather than a traditional ATS with AI features added on. The AI runs the workflow; the database and UI are designed around it.
Applicant tracking system (ATS)
Software that stores candidates and tracks them through hiring stages. A traditional ATS records the process; an agentic ATS also runs it.
Tally
The AI inside Talara: a single assistant with more than 190 tools that drafts roles, sources, screens, schedules, and builds offers, always behind a human approval. Meet Tally.
Human-in-the-loop
A model where a person is accountable for every consequential decision even when software does the work. In Talara it is enforced as a database constraint. More.
Model Context Protocol (MCP)
An open protocol for exposing software tools to AI clients. Talara is MCP-native, so its 190+ tools can be driven from a Claude conversation or any MCP client. More.
Audit trail
An immutable, attributable record of every meaningful action, carrying the actor, the subject, and the before-and-after change. More.
Fit score
An AI-generated estimate of how well a candidate matches a role, shown alongside the reasoning behind it so a recruiter can judge it rather than trust it blindly.
EU AI Act (hiring)
EU regulation that classifies AI used to recruit or select candidates as high-risk, requiring human oversight, transparency, bias management, and record-keeping. More.
See the termsin practice.
Talara is the agentic ATS these definitions describe.