This Germany-Based AI Startup is Developing the Next NLP and Open-Source Powered Enterprise Search Engine
Pitch your startup story at [email protected] Please don't forget to join our ML Subreddit
Data is everywhere. However, having access to data does not always mean having access to relevant and contextualized information to explore and extract information. Finding the correct information in the midst of a sea of texts is becoming more and more difficult.
Natural language is the most adaptable and powerful approach to communicating with data and software.
Deeply, a German startup, is working to add to natural language processing by integrating a layer of language awareness into the company’s technology stack, allowing users to access and interact with data using of language. Its flagship product, haystackis an open-source NLP framework that allows developers to build pipelines for a variety of research use cases.
Haystack-based NLP is typically implemented on a text-based database like Elasticsearch or Amazon’s OpenSearch branch and then connects directly to the end-user application via a REST API. It already has thousands of users and over 100 contributors. It uses transformer models to allow developers to create a variety of applications, such as production-ready question answering (QA), semantic document search, and synthesis. The company also introduced deep cloudan end-to-end platform for integrating custom, high-performance NLP search systems into your application.
Deepset aims to bridge the gap between research and industry by enabling developers to build flexible and powerful neural search engines that can query all kinds of data. They are developing a semantic layer for the modern technology stack, powered by cutting-edge NLP and open source.
The Berlin-based company raised $14 million in Series A funding led by GV, the venture capital arm of Alphabet. The company aims to use these funds for product development and to expand its go-to-market strategy. The company’s upcoming technological advancements include, among others, native support for voice-based search.