This video accompanies the following post: https://blog.baeke.info/2023/03/16/pinecone-and-openai-magic-a-guide-to-finding-your-long-lost-blog-posts-with-vectorized-search-and-chatgpt/
In it, we store blog posts as vectors in Pinecone, a vector database. Vectors are created with the OpenAI embeddings API. When the user asks a question, we vectorize the query string and find the closest matching vector in Pinecone. We then grab the blog post via its URL and feed the original query and post to the OpenAI ChatCompletions API.
Github: https://github.com/gbaeke/gpt-vectors
#openai #chatgpt #vectorization #embeddings