Agentic AI & MCP Specialist
Agent loops, tool calling, the Model Context Protocol, evaluation and safety — the skill set behind every AI-agent role.
Built for the fastest-growing role family in tech. Tests whether you can design and ship reliable AI agents: the agent loop and when to use a workflow instead, tool and schema design, the Model Context Protocol (MCP) and its primitives and transports, context management and RAG, prompt-injection defences, evaluation and multi-agent orchestration.
What it measures
Agent Architecture
Agent loops, workflows vs agents, ReAct, budgets and stopping.
Tool Calling
Tool schemas, argument validation, side effects and large outputs.
Model Context Protocol
Hosts, clients and servers, primitives, transports, JSON-RPC and authorisation.
Context & Prompting
RAG, system prompts, structured output and examples.
Evaluation & Safety
Eval sets, LLM-as-judge, prompt injection and human approval.
Multi-agent Systems
Orchestrator–worker patterns, cost and when not to use them.
Who it's for
- Developers building LLM agents and integrations
- ML and AI engineers
- Technical product managers working on AI features
- Learners from the ARISE AI Agents & MCP course
How the test works
- 20 questions drawn at random from a larger bank, balanced across every skill area above.
- Answer options are shuffled for each candidate, and the timer keeps running once you start.
- Your answers save automatically. If your connection drops, reopen your link to continue.
- Score 70% or more to earn a verifiable ARISE certificate. You get 2 attempts.
Try a sample question
In an AI agent, what is a 'tool'?