Considering building a chatbot to serve a technical knowledge base? Learn how to build your own knowledge base without your enterprise data leaving your enterprise data warehouse (e.g., #BigQuery) or your unstructured data (in Cloud Storage) → https://goo.gle/4eQE8lt
Created a mini project for my son as a Kid’s Tutor Bot using Telegram Bot and free Gemini and deployed it to Cloud Run. Looking into upgrading it with his school modules by using RAG. Thanks to Telegram and Google’s free resources, it’s easy to explore and experiment on projects.
Google Cloud Love this post as I got my #Masters in #ComputerScience from UC San Diego specifically for understanding #DBMS and then #exploring #algorithms like #BigQuery and #innnovation at then @Apache #Solr and #Hadoop to #index #plaintext
Does the RAG reduce the need for Dialogflow in the process? Meaning we can start faster and add more ambiguous Dialogflow later.
Is this effectively turning big query into a vector database? Just a question from a non-dev lol if anyone knows I’d be interested in a light explanation ^_^
Is there a recommend tutorial for this
Very informative
👍
Very informative
This is very informative, and thanks for sharing. But at the end of the day, what was the impact of this sophisticated architecture & services in answering user questions, even for this sample application ? Can you share some metrics around that? How much did the metrics around accuracy, hallucinations, etc. improve, and what was the cost & latency to achieve it?