Google Cloud
Documents steps to set up a Google Cloud RAG engine and Vertex AI Search with Dialogflow CX, including trial credits and regional agent creation.
What this file does
Documents steps to set up a Google Cloud RAG engine and Vertex AI Search with Dialogflow CX, including trial credits and regional agent creation.
When to use it
- Starting a RAG project on Google Cloud Vertex AI
- Setting up a conversational agent with Dialogflow CX
- Exploring Vertex AI Search pricing and trial credits
- Configuring a custom agent in a specific region like Montreal
Assumes this stack
Google Cloud
- See tracking https://github.com/ObrienlabsDev/blog/issues/47
RAG Engine
- Europe-west3 ff rag engine https://docs.cloud.google.com/vertex-ai/generative-ai/docs/rag-engine/rag-overview
- https://github.com/ObrienlabsDev/blog/issues/47
- https://github.com/GoogleCloudPlatform/generative-ai/tree/main/gemini/use-cases/retrieval-augmented-generation
- https://github.com/facebookresearch/faiss
- https://github.com/vllm-project/vllm
- https://docs.cloud.google.com/kubernetes-engine/docs/tutorials/serve-llama-gpus-vllm
Vertex AI Search (part of Gen App Builder)
Vertex AI Search is a part of Vertex AI Agent Builder
- https://cloud.google.com/enterprise-search?utm_source=chatgpt.com&hl=en
- https://cloud.google.com/generative-ai-app-builder/pricing?hl=en
- Create a project and get $1000 US of credits for 1 year - make sure to attach a billing id that does not have a pervious credit for...
- "Trial for Gen App Builder" - Following SKUs. Vertex AI Search: Search API Request Count - Standard (services/74B1-77CF-C302/skus/BADA-EE26-7BDA) Vertex AI Search: Search API Request Count - Enterprise (services/74B1-77CF-C302/skus/93D6-7280-CF05) Vertex AI Search: Search API Request Count - LLM Add-on (services/74B1-77CF-C302/skus/151B-D7A2-B5B5) Vertex AI Search and Conversation: Data Index (services/74B1-77CF-C302/skus/BC7D-6A97-90F8) Audio interactions with Vertex AI Conversation - Voice Experience (services/FBC0-AA4A-C89A/skus/1A04-3B12-74DC) Text interactions with Vertex AI Conversation - Chat Experience (services/FBC0-AA4A-C89A/skus/90FE-6D2B-0AE1)
Navigate to your gen app builder project
- TODO: determine regional resources - for PII and PBMM data in the vector store
- https://console.cloud.google.com/gen-app-builder/engines?referrer=search&project=gen-app-builder-old&supportedpurview=project
- conversational agent
- https://conversational-agents.cloud.google.com/projects/gen-app-builder-old/enable?targetUrl=%2Fprojects%2Fgen-app-builder-old%2Flocations%2Fglobal%2Fagents%3Fdialog%3DagentCreationOptions
-
- todo: introduce RAG document additions and embeddings vector store
References
What's inside
2 main sections (RAG Engine, Vertex AI Search), 1 subsection, 8 screenshots, 6 links, 1 todo item
Change this for your project
- Replace
gen-app-builder-oldwith your own Google Cloud project ID - Replace
northamerica-northeast1with your desired region - Replace the screenshot image URLs with your own captures
Where it goes
Reference documentation for a retrieval pipeline. Keep with the ingestion or retrieval code it describes.
Worth borrowing
- Using prebuilt Dialogflow CX components like the feedback flow to speed up agent creation
- Linking to GitHub issues for tracking implementation progress
Related Documents
SUMMARY
Proposes three on-prem AI architectures, modular, hybrid, and fully local RAG, with hardware specs and vendor lists.
Retrieval & Prompts
Explains how CharMemory's extraction prompt and Vector Storage settings determine memory retrieval quality in SillyTavern.
App Review Support Guide — Switch2Go
Explains an AAC app's accessibility permissions, hardware needs, and reviewer walkthrough to pass App Store review.
RFC-BLite: High-Performance Embedded Document Database for .NET
Specifies an embedded document database for.NET with zero-allocation I/O, C-BSON format, and ACID transactions.