Combining Intent-Based AI with Agentic AI in Cognigy

Discover the real difference between Cognigy Tokens and Cognigy Script! Tokens are quick UI shortcuts, while Script offers full flexibility for inserting variables, logic, math, and more. Learn when to use each, how to create custom Tokens, and why mastering Script can supercharge your bot building. Perfect for beginners and pros alike!

In Cognigy, an intent score like 0.90 reflects how confidently the NLU model matches a user’s input to a trained intent. Behind the scenes, it's likely the output of a sigmoid function from logistic regression — a probability-like measure, but not a guaranteed truth. Understanding this score helps you set smarter thresholds, design better conversations, and avoid overtrusting machine confidence.

Learn how leading firms like McKinsey empower employees with AI-driven knowledge bases. In this guide, we show you how to build a dynamic assistant using Cognigy.AI, OpenAI embeddings, Amazon S3, and Pinecone—giving your team instant access to case studies, frameworks, and critical data, faster and smarter.

In Cognigy AI, Lexicon and Knowledge serve distinct purposes in natural language understanding. Lexicon is rule-based, using hardcoded synonyms to normalize input before intent recognition — no word vectors involved. In contrast, Knowledge leverages semantic search with vector embeddings to retrieve answers from documents, making it ideal for dynamic Q&A use cases. Knowing when to use each unlocks smarter virtual agents.

According to McKinsey, employees spend an average of 1.8 hours per day —or 9.3 hours per week —seeking internal support. Depending on the...

How long does it take to learn Cognigy AI? 100 hours? 200 hours? More?