Weaviate, a vector database provider, has announced the addition of three new agents to its development stack, designed to facilitate the development of generative AI-based applications. These agents, which are modular agentic workflows, utilize large language models (LLMs) and prompts to interact dynamically with data in Weaviate.
The three agents showcased by Weaviate are the Query Agent, Personalization Agent, and Transformation Agent. According to Noel Yuhanna, principal analyst at Forrester, these agents could help developers automate their processes for building AI applications and enable natural language interactions with data. The agents are expected to be made available as a separate service, allowing developers to simplify complex query workflows, organize datasets, and personalize agent behavior.
The Query Agent, currently available in Public Preview, is aimed at helping developers simplify complex query workflows and boost RAG pipelines by querying data using natural language. This eliminates the need for constructing a query understanding pipeline, which Weaviate claims is challenging to build, maintain, and requires specialized expertise. Instead, the Query Agent uses function calling, which structures queries using predefined function calls in JSON format, with optional arguments for search, filters, aggregation, and grouping.
The Transformation Agent, on the other hand, allows developers to augment and organize datasets with a single prompt. Use cases include cleaning and organizing raw data for AI, generating and enriching metadata, automatically categorizing, labeling, and preprocessing data, or even translating the entire dataset. The Personalization Agent is targeted at personalizing agent behavior, enabling developers to tailor the agents to their specific needs.
Industry analysts believe that Weaviate's move is in line with the growing trend of integrating agents to automate various functions and processes to accelerate AI workloads. "The advantages of these agents are that they can evolve independently, making it easier for enterprises to adopt them quickly than committing to a single, all-encompassing AI solution," explained Yuhanna.
However, Bradley Shimmin, global technology analyst at The Futurum Group, notes that Weaviate is not the first vendor to use generative AI to augment and automate tasks for developers, database administrators, data scientists, and even business users. Shimmin views these agents as clever marketing, with Weaviate being the first vector database provider to market agents as a family of developer-focused tools.
The release of Weaviate agents comes at a time when the database landscape is rapidly changing, with a growing number of vector databases to choose from, particularly among stand-alone players like Weaviate, Qdrant, Faiss, Chroma, and Milvus. Shimmin believes that Weaviate's move may help the company compete somewhere between stand-alone and traditional database offerings, such as PostgreSQL, Oracle Database, or Microsoft Cosmos DB.
Weaviate's target customers for these agents are startups and small and medium enterprises. During the Preview phase, the agents will be free to use as part of Weaviate Serverless Cloud and its free developer sandbox, with detailed pricing to be updated in the future.
In conclusion, Weaviate's introduction of AI-powered agents marks a significant step forward in simplifying AI application development and data interaction. As the database landscape continues to evolve, it will be interesting to see how Weaviate's agents impact the industry and how the company plans to establish a rich developer ecosystem around these agents.