Enso Technologies Ltd. today launched a marketplace that will enable companies to access artificial intelligence agents for a monthly subscription.

The platform is the fruit of a partnership with LangChain Inc., a fellow startup that provides tools for building AI applications.

Tel Aviv-based Enso launched last July with $6 million in initial funding from Google Research head Yossi Matias and other investors. The company is led by Chief Executive Officer Mickey Haslavsky, who previously co-founded developer tooling provider RapidAPI. The latter company raised more than $270 million in funding before being acquired by Nokia Corp. last year. 

Enso’s newly launched marketplace provides access to more than 300 AI agents, or AI applications that can perform business tasks with a high degree of autonomy. Some target common AI use cases such as generating marketing copy and drawing visual assets for websites. Others focus on automating industry-specific tasks for companies such as car dealerships.

The company says some of the AI agents on its platform can also perform complex data processing tasks. Salespeople, for example, can use them to automatically find leads in a large database of potential customers. It’s possible to narrow down the results based on the industry where a lead is active, company size and other parameters.

Companies can access the agents in Enso’s marketplace through a $49 monthly subscription. That price point is designed to make the platform accessible for small and midsize businesses, which usually lack the resources to build their own AI agents.

For AI agent developers, in turn, Enso positions its marketplace as a way to more easily reach potential customers. The platform automates some of the administrative tasks involved in monetizing AI software. That includes payment processing and customer service. 

Enso is launching the marketplace in partnership with LangChain. The venture-backed developer tooling provider provides an open-source framework likewise called LangChain that eases the task of integrating large language models into applications. It includes prepackaged code components and connectors that help LLMs retrieve data. 

The partnership with Enso focuses not on LangChain’s namesake framework, but rather a second product from the company called LangGraph. That tool is likewise designed to ease the development of LLM-powered applications. However, it provides a more advanced set of features than LangChain.

LangGraph allows developers to build applications that include multiple AI agents. Those agents can be equipped with memory. According to LangChain, that enables LLM-powered applications to take into account past prompts from a user when processing a new query. 

LLMs often process prompts in multiple steps. LangGraph’s memory feature enables users to pause the processing workflow after each step to check the result and provide a correction if necessary. To streamline the task, developers can create safety guardrails that filter harmful output. 

Developers can deploy LangGraph-powered agents to Enso’s marketplace in a few clicks. “Our mission at LangChain has always been to empower developers to build powerful applications with AI agents,” said LangChain Chief Executive Officer Harrison Chase. “This partnership with enso takes it a step further — developers now have an effortless way to reach millions of small businesses and generate recurring income.”

Image: Enso