How we did it: Enhancing our search engine with AI and vectors

We built a cutting-edge hybrid search engine combining vector and keyword search for enhanced relevance and performance. We started by scraping the relevant sites, creating a vector index, building an inverted index for keywords, and implementing both vector and keyword search individually. To combine these two search methods, we implemented re-ranking using reciprocal rank fusion, which combined the strengths of each retriever into one final results list. The front-end was integrated with the existing website, featuring an initial modal that popped up with the top five results and a button to see the rest. Our next steps include improving search performance by returning snippets or finding more efficient ways to scrape the websites, and using our search engine as the source for an AI chatbot for a conversational experience.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.