BlazAl Technology
Last updated
Was this helpful?
Last updated
Was this helpful?
The conversational AI, BlazAl, is built using:
Conversational AI Core:
Rasa: Open-source machine learning framework used for building the core conversational AI, handling dialogue management and natural language understanding (NLU).
Spacy: Employed for its efficient Natural Language Processing (NLP) pipeline, assisting in tasks like entity recognition and intent classification.
DIETClassifier (Dual Intent and Entity Transformer): Rasa's architecture component used for joint intent classification and entity extraction, crucial for understanding user commands accurately.
Knowledge Augmentation:
Vertex AI: Google Cloud's AI platform is leveraged to answer questions or handle queries that fall outside the scope of BlazAl's internal Large Language Model (LLM) capabilities, ensuring comprehensive user support.
Backend & Frontend:
Node.js: Used for the backend infrastructure, handling API requests, business logic, and integration with blockchain components.
ReactJS: Powers the frontend user interface, providing a responsive and interactive chat experience.