Weaviate Cloud Shared

What do you like best about the product?

I like that Weaviate makes it fairly straightforward to work with vector data and search through it. The documentation is useful when setting things up, and the dashboard makes it easy to keep track of the projects and collections. It also works well for experimenting with AI and search-related use cases.

What do you dislike about the product?

It can take some time to understand the different concepts if you're new to vector databases. Some of the configuration options can also feel a little overwhelming at first.

What problems is the product solving and how is that benefiting you?

Weaviate helps with storing and searching vector data, which is useful for AI-based search and retrieval use cases. It gives a convenient way to experiment with semantic search without having to build the whole infrastructure from scratch.

What do you like best about the product?

Weaviate is an all-in-one platform for building vector search, RAG, and agent memory management for us. With it, we design, build, and ship the entire AI stack, from local development through to our AWS production environment. Above all, it’s open source, which helps eliminate vendor lock-in concerns for our organization and also provides the flexibility to customize, improving overall performance.

With its vector database, we store, index, and retrieve different types of media information for our products, which supports scaling AI agentic systems. It also includes an Explorer that runs semantic, keyword, and hybrid search with aggregation, without needing to write GraphQL that saves time for our engineers.

Weaviate embeddings also help deliver efficient, faster models like Snowflake designed for enterprise-level retrieval operations.

What do you dislike about the product?

Although it has decent features, the UI feels slightly outdated based on my experience. When it comes to integrations with third-party platforms outside of the machine learning ecosystem, it’s not quite there yet. That said, because of the open-source community on GitHub, I expect the number of integrations to grow over time for other tech stacks that we use daily.

It does offer a 7-day free trial, but after that, calculating the overall monthly cost is complicated. It charges different services at different rates, which makes it hard to understand the final price per month. We have to use a calculator to add up all the final costs.

What problems is the product solving and how is that benefiting you?

Its native data query agent turns natural language–based questions into database queries and operations, which eliminates the time we used to spend writing SQL. This AI-based query agent delivers good results with dynamic filters, cross-collection routing, and source citation for our conversational AI apps, including customer-support chatbots. We can upload or add new databases from our knowledge base into collections, where we’re able to review detailed metadata and properties to evaluate the dataset.

It also includes a fully managed memory for AI agents called Engram, which remembers personalized preferences and decisions across all of our agent sessions. This helps shrink the context window and sends relevant, structured memories to our production agents during customer interactions.

We noticed our database costs dropped by 11% thanks to more efficient resource consumption and an optimized memory footprint. On top of that, it offloads tenant details to cold storage, which further reduces storage usage and cost.

It also secures and isolates our customer data, which is an important safety net for us. The open-source community has created support documentation that’s handy when we need to troubleshoot and fix issues.

What do you like best about the product?

I use Weaviate for efficiently storing vector data, which is crucial for embeddings and the retrieval phase of a RAG implementation. I appreciate its open-source nature and the efficient vector retrieval, which is necessary for ensuring the retrieval phase is accurate and fast. Additionally, AWS support is great, and the initial setup was straightforward.

What do you dislike about the product?

I think finetuning the self-hosting option is a bit complex. The RAM usage can be high depending on the dataset, so if self-hosting, several decisions regarding this have to be made. You have to scale horizontally in some cases if you want to keep it running smoothly. The initial setup was great, but finetuning is not as easy.

What problems is the product solving and how is that benefiting you?

I use Weaviate for efficient vector data storage in a RAG pipeline. It enhances the retrieval phase by being fast and accurate, which is essential as the pipeline is large and can't afford extra latency.

What do you like best about the product?

Using weaviate, it makes it straightforward to build semantic search and retrieve relevant information without having to manage vector search logic from scratch

What do you dislike about the product?

I felt like there was higher memory consumption in the case of weaviate

What problems is the product solving and how is that benefiting you?

it mainly solves the problem of storing and searching vector data for semantic retrieval

What do you like best about the product?

What I like most about Weaviate is how easy it is to build semantic search and RAG applications around it. The vector search is fast and flexible, and I like that it supports structured metadata along with embeddings, so I can narrow down results without making the retrieval logic overly complicated.

What do you dislike about the product?

The main thing I find challenging with Weaviate is that there can be a learning curve when setting up more advanced configurations, especially around schemas, indexing, and tuning retrieval. It also takes some time to understand how to get the best results from the vector search instead of relying on the default setup. For smaller projects, it can sometimes feel like more infrastructure than I actually need.

What problems is the product solving and how is that benefiting you?

Weaviate mainly helps with the search and retrieval side of AI applications. I use it to store and search vectorized data based on semantic similarity, which is much more useful than relying only on keyword matching. It makes things like RAG and knowledge-base search easier to build, and helps return more relevant context to the application without having to manually manage the retrieval layer.