Course Overview
Your team wastes hours digging through PDFs, FAQs, and spreadsheets just to answer one question?
Here's the truth: You don't need developers or coding skills to fix this. This two-part course teaches you to build a complete RAG (Retrieval-Augmented Generation) system in Flowise that reads YOUR documents—PDFs, Word files, CSVs, even websites—and answers questions with pinpoint accuracy grounded in your actual business data, not generic internet information.
Who Is This Course For?
What to Expect
Curriculum
4 modules • Lifetime Access
Learn RAG's two-phase workflow (Indexing + Retrieval), set up Flowise Document Store, configure Document Loaders for FAQ files, and apply Recursive Text Splitters with 1000-char chunks and 200-char overlap
Objective: Transform raw business documents into properly chunked, AI-ready knowledge segments ready for vectorization
Value: 100 HKD
Convert text chunks into1,536-dimension embeddings via OpenAI's text-embedding-3-small, store vectors in Pinecone, then configure PostgreSQL Record Manager with namespace matching and cleanup modes (Incremental, Full, None)
Objective: Build semantic search that finds answers by meaning—not keywords—with zero duplicate data and sustainable storage costs
Value: 500 HKD
Assemble Tool Agent + ChatOpenAI (GPT-4.1) + Buffer Memory + Document Store Retriever into one chatflow, fine-tune Top K (4–6) and Temperature (0.2–0.7), then deploy via embed widget, API, or share link
Objective: Launch a production-grade assistant that answers real queries with source attribution, conversation memory, and edge-case handling
Value: 300 HKD
Value: 100 HKD
4 modules of action-packed content
- Transform raw business documents into properly chunked, AI-ready knowledge segments ready for vectorization.
- Build semantic search that finds answers by meaning—not keywords—with zero duplicate data and sustainable storage costs.
- Launch a production-grade assistant that answers real queries with source attribution, conversation memory, and edge-case handling.
Total value: 1000 HKD
You Pay = Only 498 HKD