Announcing Our New Funding Round Led by Balderton Capital

As LLMs continue to proliferate and vector search is taking off, we’re thrilled to announce this next series of funding.

Abstract image that reads "deepset RAISES $30M."

We have some exciting news: we just raised $30 million in fresh funding! As large language models continue to capture the collective imagination and vector search is taking off, we’re thrilled to announce this next series of funding, which is led by Balderton Capital with the participation of existing investors GV (Google Ventures), Harpoon, System.One, and Lunar.

Organizations across industries are in a race to integrate LLM features into their products. Our LLM platform deepset Cloud is the world's first model-agnostic developer platform for AI teams looking to build customized, flexible, yet robust systems powered by the latest language models. A SOC 2-certified platform that lets its users retain full ownership of their data, deepset Cloud allows organizations to unlock the value of LLMs, get to market faster, and thereby gain a competitive advantage over their peers.

deepset has been at the vanguard of applying LLMs to practical use cases since the company’s inception five years ago. From day one, we worked with major enterprise customers to build bespoke NLP solutions for their real-world use cases. We soon turned our learnings into the acclaimed open source framework Haystack, later adding the commercial offering deepset Cloud to our product portfolio.

With Haystack, our core technology, developers can choose from a wide selection of components required in a modern NLP project: proprietary and open source LLMs, vector databases, file converters, text embedding models, and much more. These components are plugged into pipelines or agents to build anything from a Google-like search engine on company documents, to conversational AI, to powerful information discovery systems.

Last year, we raised series A funding with the goal of implementing deepset Cloud as a platform that does the heavy lifting required to get an AI-powered project off the ground. We’re proud to say that we have managed to build the product we envisioned – thanks to our prior experience helping many different teams build and deploy real-world LLM applications. 

With deepset Cloud, AI teams can focus on building for production rather than being held up by implementation details. It covers the entire life cycle of a modern NLP application – experimentation, production, and observability – while making it easy to compare and exchange different language models: a must in the fast-paced world of LLMs. The platform allows different product team stakeholders to work in a unified environment with fast prototyping, frequent feedback cycles, and easy customization.

Our clients come from fields as diverse as online media, legal, finance, and insurance. They praise deepset Cloud’s emphasis on usability and convenience, which enables users to compare different complex setups side by side, run an experiment at the click of a button, scale hardware automatically to growing or shrinking traffic, showcase a project prototype to different stakeholders, and gather feedback – all within one interface. 

The money raised will enable us to consolidate our status as the leading LLM platform by further developing our offering for our customers around the globe. We’ll do this by refining deepset Cloud for RAG (retrieval-augmented generation) applications: in particular, improving the evaluation of every component in a RAG pipeline. We’ll also focus on making the platform viable for customers with heavy privacy constraints by optimizing for virtual private cloud (VPC) setups. Finally, we’ll home in on diversifying and improving LLM observability in deepset Cloud, so that customers can be maximally confident in their application’s performance in a production environment.

We have come a long way since our early days, and our company has grown considerably in the past year. But despite doubling in size in such a short time, the team has managed to maintain a closely-knit feel. deepset is now a 50-people-strong powerhouse made up of knowledgeable, talented, empathetic, and ever-curious individuals dedicated to collaborating in the best way possible. Creating such a supportive and productive atmosphere is perhaps our greatest accomplishment, and one that we attribute to each and every team member 🙌!

In addition to developing our commercial offering, we remain deeply committed to Haystack, our flagship open source framework, and we would like to thank our wonderful community of NLP enthusiasts, active collaborators, and attentive observers on GitHub for making this project possible in the first place 💙.

By the way: we’re hiring! Want to help organizations tap into the enormous potential of today’s most exciting language technologies? Have a look at our job openings and join us as we take our next, exciting step.