AI, Skills & Career Terms Explained

11 of 11 terms

Agentic AI

AI systems that plan multi-step work and use tools to complete it, rather than only answering a single prompt. At Prakura, agentic actions with real-world effect require human approval.

AI Operations Center →Agent Studio →

AI Agent

A software assistant with a defined role, knowledge sources, tools and permissions. For example, a QA Agent drafts test scenarios but cannot commit code.

Agent catalog →

AI Copilot

An assistant that works alongside a person inside their tool — suggesting, drafting or explaining — while the person stays in control.

HR Assistant →

RAG (Retrieval-Augmented Generation)

A technique where the AI first retrieves passages from authorized documents and then answers using only those passages, citing them as sources.

Knowledge Agent →Data sources →

MCP (Model Context Protocol)

An open protocol for connecting AI applications to tools and data sources through a standard interface, so integrations can be reused across assistants.

Integrations →

Placement Readiness

A view of how prepared a student is for a specific placement drive, based on skill evidence, assessments and mock interviews — not a prediction of selection.

Placement Engine →

Career Passport

A student-owned profile of verified skills, projects, coding evidence and readiness that can be shared selectively with recruiters.

Placement Engine →

Skill Graph

A connected map of skills linked to courses, roles, projects and evidence, used to explain recommendations and show gaps.

Intelligence Graph →

Digital Twin

A simulation of an organisation — such as a campus or workforce — used to explore "what if" scenarios before making real changes.

Digital Twin demo →

Human-in-the-loop

A design pattern where a person reviews and approves AI output before it takes effect.

AI Trust Center →

Groundedness

The degree to which an AI answer is supported by its cited sources. A key metric for knowledge assistants.

AI evaluation →