Rasa
Rasa is an open-source framework for building Conversational AI – with NLU, Dialogue Management, and integrations for enterprise chatbots.
Rasa is the leading open-source framework for enterprise chatbots – with NLU, Dialogue Management, and full data control on-premise.
Explanation
Rasa consists of NLU (Intent + Entity Recognition), Core (Dialogue Management via Stories/Rules), and Action Server (Custom Backend Logic). It runs on-premise and offers full data control.
Marketing Relevance
Standard framework for enterprise chatbots with data privacy requirements. Alternative to cloud services like Dialogflow or Amazon Lex.
Example
A bank uses Rasa on-premise for a chatbot that retrieves account balances, initiates transfers, and schedules appointments – without sending data to cloud services.
Common Pitfalls
Steep learning curve. Training data must be manually created. Scaling requires Kubernetes infrastructure. LLM integration still in development.
Origin & History
Founded 2016 in Berlin. Rasa NLU (2017) started as open-source intent classifier. Rasa Core (2018) added dialogue management. Rasa 3.0 (2022) brought transformers. CALM (2024) integrated LLMs for more flexible dialog design.
Comparisons & Differences
Rasa vs. Dialogflow
Dialogflow is Google Cloud-based and simpler; Rasa is open source, on-premise, and more flexible but more complex.
Rasa vs. LangChain
LangChain orchestrates LLM chains; Rasa is a complete chatbot framework with NLU, DM, and action server.