Building your own Multiskill AI Assistant with DeepPavlov
summary
Concrete claim: DeepPavlov provides an open source conversational stack that runs a neurosymbolic architecture combining neural annotators and symbolic dialogue flows to build a multiskill assistant. It ships a Dream distribution, a Python DSL (DFF) for dialogue flows, and tooling to run locally or via proxy. The demo shows 40 skills and ~20 annotators, sustains 7 rps in competition and estimates 100k rps on Kubernetes/AWS, with runnable Postman/console APIs and visual flow editing.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 8:19Augmented intelligence history
- 17:03DeepPavlov Agent overview
- 24:25Building skills with DFF
- 30:05Knowledge graphs and slot filling
- 31:29Open-source AI assistant opportunities
- 35:20Q&A
related talks
33mBuilding the next-generation Conversational AI with Python and Deep Learning
34mBuilding Reliable AI Assistants with Small, Self-Hosted Language Models
30mMindmeld: An Open-source, Deep-domain Conversational AI Toolkit
01h 29mGoing Beyond “Sorry, I Didn’t Get That”: Building AI Assistants that Scale...
30mConversational AI: Building clever chatbots
46mBeyond Deep Learning: The Cognitive AI Frontier
31mDSPy: The End of Prompt Engineering as We Know It
41mGoing beyond copilot with AI Agents