8Knot Chatbot
summary
A chatbot for 8Knot that recommends which of the 30–40 visualizations to show and then uses LLM function-calling to populate Plotly graph parameters; graph JSON returns feed back as structured context so the model can explain trends. Retrieval uses graph name + description + vector DB of embeddings instead of slow agent reasoning. Backend experiments include breaking graph generation into AWS Lambda functions (cold starts accepted) and a responsive UI; source Augur data covers about 30,000 projects.
AI-written summary · about
related talks
29mKnowledge graph based chatbots
01h 36mTutorial: From Planni...
34mSearching for My Next Chart
28mWhat the RAG? Demystifying Retrieval Augemented Generation
33mScalable LLM Inference on Kubernetes With NVIDIA NIMS, LangChain, Milvus and Flu...
35mBuilding complex data visualisations using ChatGPT
46mExplorations in Interactive Visual Analytics
19mWisdom-Driven Knowledge Augmented Generation at Scale