Berlin-based Jina.ai, an open source startup that uses neural search to help its users find information in their unstructured data (including videos and images), today announced that it has lifted a round of $ 30 million Series A financing led by Canaan Partners. New investor Mango Capital, along with existing investors GGV Capital, SAP.iO and Yunqi Partners also participated in the round, which brings the company’s total funding to $ 39 million to date.
Jina.ai CEO and co-founder Han Xiao, who co-founded the company with Nan Wang and Bing He, explained that the idea behind neural research is to use deep learning neural networks. to go beyond traditional keyword-based research tools. Using relatively new machine learning technologies like transfer learning and representation learning, the company’s core Jina framework can help developers quickly build research tools for their specific use cases.
“Given an image, sound, video or whatever, we first use deep neural networks to translate this data format into a universal representation,” Xiao explained. “In this case, it’s mostly a mathematical vector – 100-dimensional vectors. And then the correspondence [algorithm] does not count how many letters match but counts the mathematical distance, the vectorial distance between these two vectors. That way, you can basically use this type of methodology to solve all kinds of data research or relevance issues.
Xiao described Jina as similar to TensorFlow for research (with TensorFlow being Google’s open source machine learning framework). Much like TensorFlow or PyTorch set the design blueprint for how people design AI systems, Jina wants to define how people build neural research systems – and become the de facto standard for doing it in the process.
But Jina is just one of the company’s current products. It also offers the Jina Hub, a marketplace that allows developers to share and discover the building blocks of Jina-based neural research applications, as well as the recently launched Finetuner, a tool for fine-tuning any deep neural network.
“Over the past 18 months, we’ve put a lot of effort into building the basic infrastructure, building the foundation for this great neural research tower – and that part is already done,” Xiao said. “And now we’re slowly building the first floor, the second floor of this big building – and we’re trying to provide an end-to-end development experience. “
The company claims that the Jina AI developer community currently has around 1,000 users, with apps ranging from a video game developer using it to automatically populate relevant game assets by right-clicking on several of its game publishers to a legal tech startup that uses it to enable their chatbot to deliver a question-and-answer experience that draws on data from PDF documents.
The open source Jina framework already has nearly 200 external contributors since its launch in May 2020 and the company also hosts an active Slack community around the project.
“The reasons we do open source are mainly due to the speed of open source – and I think the speed of development is a key factor in the success of a software project. A lot of software dies just because this speed reaches zero, ”Xiao said. “We are building the community and leveraging the community to gather feedback to iterate quickly. And that’s very important for infrastructure software like us. we build the community. And we’re leveraging the community to gather feedback to iterate quickly. And that’s very important for infrastructure software like us. You need all of these top developers to give your opinion on usability, accessibility, etc. in order to improve it quickly.
Jina.ai plans to use the new funding to double its team and most importantly to expand its operations in North America. With this expanded team, the company plans to invest in R&D to expand Jina’s global ecosystem and launch new tools and services around it.
“Traditional search systems designed for text data don’t work in a world full of images, videos and other multimedia elements. Jina AI is turning businesses from black and white to color, unlocking unstructured data in a fast, scalable and data independent way, ”said Joydeep Bhattacharyya of Canaan Partners. “The first applications of its open source framework are already showing signs of the future, with underlying neural research opportunities to improve decision-making, refine operations and even create new sources of revenue. “