Relevance AI, the Australia based start-up founded in 2020, has introduced a powerful developer-first vector platform to help developers do more with unstructured data and to aid data scientists rapidly experiment with vectors. Relevance AI’s products are already used by more than 3 million end users, with 100 million weekly requests across all sectors including SaaS, e-commerce, education and gaming. Today the company announced an early-stage investment led by New York-based global private equity and venture capital Insight Partners with participation from Galileo Ventures and Archangel Ventures. Relevance AI plans to use the funding to expand its global customer base and double its headcount
“Most businesses are not really utilizing vectors and only relying on structured data analysis to surface quantitative insights for decision making. However, 80% of business data is unstructured that comes in forms such as text, images, audio, user interactions and other formats” said Relevance AI’s co-founder Jacky Koh. “We’re thrilled to have secured this funding, as it will enable us to grow our team to continue pushing the boundaries of what’s possible and to build a highly reliable and valuable product for our customers. With this backing and unique perspective of the team at Insight Partners, we expect to grow rapidly over the next year.”
“Historically, only the tech giants have been able to make use of vectors in the machine learning architectures. Relevance AI enables companies of all sizes to leverage vector-based technology,” said George Mathew, Managing Director at Insight Partners. “Its strong founding team has created an advanced API powered offering which has positioned Relevance AI as the innovative, early enterprise leader. We look forward to partnering with Relevance AI as it continues to scaleup. “
Relevance AI’s goal is to empower developers with the ability to experiment and analyse vectors for unstructured data analysis and to democratise a technique that is nearly exclusively used by large tech companies such as Google, Facebook, TikTok and Spotify. Vectors are a specialized format that creates a high dimensional numerical representation of unstructured data that allows a computer to extract more semantic meaning and similarity out of it. With vectors, developers and data scientists can use it to match data on semantic similarity. Being able to determine similarity enables a wide range of applications such as semantic search, product recommendations, topic analysis and more.
For example, Google vectorizes text data from search queries and web content to create the most powerful search engine that semantically matches a text query to the most similar and relevant pages. Spotify vectorizes music lyrics, audio, and reviews to find the most semantically similar songs to the songs you listen to help create the most personalized and addictive discovery playlist.
Relevance has spent the last year building and scaling its platform with many early customers including TFTactics and Overwolf to empower them with unstructured data analysis through vectors.
TFTactics a gaming application, has been able to build an AI coach that takes unstructured game actions of their millions of users and performs unstructured data analysis such as identifying different cohort of playstyles and most similar players.
Relevance’s Discovery engine is also available to customers as a simple way to get into the world of vectors, with a specialised API simplifying the process for anyone wanting to build a more powerful set of discovery features. Guided onboarding and self-service are available now at https://relevance.ai/discovery.