NEOPOLIS AKADEMY
Website chatbot capable of reading the site and documentation
Advanced practical exercise to create a chatbot that can read a website and its documentation. Combines n8n, Perplexity and Pinecone to index content, retrieve context and reply via Gmail, in a secure, controlled setup.
View this course on Akademy ↗Enrolment and practical details are available on Neopolis Akademy.
What you will explore
Create a site chatbot that extracts and indexes pages and documentation using Pinecone.
Use Perplexity for contextual retrieval and n8n to orchestrate flows and send replies via Gmail.
Operate on fictional data with controls and human validation to mitigate risk.
STEP BY STEP
Course programme
01Website chatbot capable of reading the site and documentation
Build a chatbot that can extract and use a website’s content and documentation. The practical work explains page ingestion via pipelines (n8n), vector indexing (Pinecone), similarity queries and contextual enrichment using Perplexity and Gmail.
Objective: enable the chatbot to read, index and answer from the website and its documentation.
Prepare the environment: crawlers, embedding storage (Pinecone) and connections to Perplexity/Gmail.
Build progressively: ingestion, chunking, index creation and similarity queries for precise answers.
Security checkpoint: access rights management, index updates and limiting sensitive data indexed.
Explore this module on Akademy ↗Programme source: Neopolis Akademy. Original course page