The team has been hard at work over the summer and we are releasing all our most requested features all at once.
Users asked to make it easier and faster to build complete RAG applications, so we added our chat agent, which is fully hosted, no-code agentic chatbot connected to your RAG pipeline with a long list of features (rich authentication, multiple languages, custom branding, and much more). We also made it super easy to add this agentic chatbot to your website with our agent widget using a single line of code.
We've supported local MCP for a while, but our users were asking for remote MCP support with API key and CORS support, so we built that to make it easy to use your RAG pipeline with Claude, Cursor, and a host of other MCP compatible clients.
For time sensitive use cases, our users where asking for pipelines that sync up with the source data as soon as it is available. So, we added real time pipelines, which are always running and syncing the source data to the vector database.
And to improve how AI agents (including the chatbots and widget) get the data they need to do their job, we've added hybrid search, advanced metadata filtering, and knowledge graph capabilities to our pipelines.
We are excited to get your feedback on these Vectorize 2.0 features. Let us know what you think!
building rag pipelines with no code sounds amazing :)) the hybrid search and knowledge graph features could really improve retrieval accuracy
impressive how you've made rag pipelines so accessible - the no-code approach is a game changer for developers :))
I love how Vectorize focuses on keeping data up-to date that’s something that often gets overlooked. Does the refresh pipeline kick in automatically when there are updates to the source or does it follow a set schedule?
Streamlined RAG pipelines are just what teams need how long does it take to set up?
Vectorize 2.0 brings it all together, hosted chat agents, no-code chatbot builder, one-line widget integration, Remote MCP for Claude & Cursor, real-time pipelines, and smarter retrieval with hybrid search + knowledge graphs. It’s everything users asked for, now in one powerful stack.
How closely is Vectorize linked to a specific vector database? Can it operate across platforms like Pinecone, Weaviate, Milvus, etc. or is it tailored for just one environment?
Finally RAG and evaluation all in one. Which vector databases are ready to go right out of the box?
Congrats on the launch! Here's my two cents:
Great promise to liberate developers from 'toy' AI projects by generating production-ready, open-stack code (React/Node/Postgres) that prevents vendor lock-in and ensures scalability.
While I think primary risk is that the definition of 'production-ready' is subjective and enterprise-grade requirements around security, observability, and compliance may render the generated code insufficient, positioning the tool as just another boilerplate generator in a crowded market.
If so, I'm thinking would it be the first priority to validate the definition of 'production-ready' with enterprise architects? Focus the initial wedge on technical founders in who prioritize long-term portability over the immediate convenience of closed ecosystems.
Love how you bundled the most requested features together—this looks really useful!
Hello, your chatbot has an excellent feature. I am amazed with its working style. Thank you so much for your creation.
Big release, love how Vectorize combines hybrid search + real-time pipelines with a no-code agent widget. Curious, how does the knowledge graph layer impact retrieval speed under load?
