Understanding AI Labs
AI Labs is a section within the Bridged dashboard where you create and manage AI Engines.
Engines are the behaviour layer behind your agents. They define how an agent responds, what tone it uses, which rules it follows, how it detects intent, and what it should do when it cannot answer confidently.
Use AI Lab when you need to create, review, adjust, or improve the behaviour of an agent.
What is an Engine?
An Engine is a set of instructions and guidelines that controls how an agent behaves.
It can define:
personality and tone
response structure
intent detection
content generation rules
fallback behaviour
brand and quality guidelines
Think of an Engine like a recipe: it tells the agent how to process information and generate responses in a specific, controlled way.
This helps keep agent behaviour consistent across events, brands, channels and workflows.
What is Fine-tuning?
Fine-tuning is the process of improving an Engine after it has been launched.
It uses real user feedback, such as star ratings, comments, and manual response edits, to help create a more accurate and useful version of the Engine over time.
What you can do in AI Labs
Create and manage Engines
Build custom Polling or Knowledge Engines from scratch or use system defaults.
Engines control how agents respond, helping you keep tone, behaviour and brand alignment consistent.
Fine-tune Engines
Review feedback from live agent responses and use it to create improved versions of an Engine.
This helps your agents become more accurate, useful and aligned with real user needs over time.