Algolia acquires to allow users to “search as they think”

Data search platform Algolia announced the acquisition of, a San Francisco-based neural search startup.

Algolia is known for its keyword-based search and discovery platform that processes trillions of queries from hundreds of global data centers and is “API-first” or optimized for client applications.

Search.ioThe core product of is a vector search engine called Neuralsearch which is built on an AI algorithmic processing engine that uses neural hashing on top of vectors. Algolia will combine its keyword search and Neuralsearch into a single API.

Algolia says it will stay true to its mission of making search and discovery intuitive, fast, and scalable with affordability and ease of use in mind. The company says the acquisition will enable it to provide the first and only API-based search and discovery platform with a hybrid search engine consisting of both keyword search and semantic search in a single API.

“Our mission, vision and purpose fuel discovery. So far, we’ve done it largely with keyword research. With the addition of’s vector search engine, we are going to significantly disrupt the search market,” said Bernadette Nixon, CEO of Algolia. “We will be the only product on the market that combines keyword search with vector-based image and semantic search, as well as vector-based recommendations. Vendor consolidation is back in vogue, and being able to get best-in-class capabilities from a single vendor is powerful in today’s economic climate.

Vector search is a technique in which an AI engine attempts to match an input term to what is called a vector, or an array of features generated from a catalog of objects. Features are derived from catalog objects through a machine learning model that converts those object features into a two-dimensional vector with up to hundreds of dimensions.

However,’s Neuralsearch is a bit more complex than your typical vector search, as it uses vector search in combination with hashes created by a neural network. The company claims that this technique provides faster and more accurate results than vector search alone. Hash is a type of data retrieval that relies on the statistical properties of the interaction of keys and functions and is valued for its efficiency in computing and storage resources. In a blog CEO Hamish Ogilvy explains how neural networks can optimize hash functions which, compared to an original vector, can retain near-perfect information much faster and with a smaller storage footprint.

This example from website explains how Neuralsearch understands a customer’s intent when searching for a product. Source:

“Artificial intelligence was built on the back of vector arithmetic. Recent advances show that for some AI applications, this can actually be significantly outperformed (memory, speed, etc.) by other binary representations (such as neural hashes) without significant compromise in accuracy,” Ogilvy said. “Once you work with things like neural hashes, it becomes apparent that many areas of AI can transition from vectors to structures based on hash and trigger a huge acceleration of AI advancement.”

The combination of keyword research and vector semantic search from Algolia and will allow users to search for either specific keywords or natural human expressions, which Algolia says offers users the ability to “search as they think”. Additionally, the company says it is tackling the problem of long-tail search queries, or those keyword phrases that tend to be longer and very specific and are associated with customers who are on the point of making a purchase. These queries have traditionally been more difficult to accommodate.

“Industry-wide, retailers are leaving money on the table because it’s hard to generate revenue from long-tail search queries (such as ‘beautiful fall outfit for mother of the bride’ “), which could potentially represent up to 55% of all search queries today,” Nixon noted. satisfied with less popular or less sought after products.Our new Algolia hybrid search engine solves this long tail problem – truly putting search on autopilot at a price 90% less than other vector search options.

According to the CEO of Ogilvy, all employees have been offered new positions within Algolia, reflecting Algolia’s current hiring frenzy. The company says it created more than 145 new jobs in the second quarter and doubled its number of employees over the past year.

“Welcoming the team and launching our hybrid search engine represents the start of Algolia’s next chapter,” said Nixon. “Integrating vector search into our keyword research provides us with a game-changing launching pad to solve more of our customers’ search and discovery challenges and deliver a price advantage they won’t see elsewhere.”

“We are thrilled to join a world leader in research and discovery,” said Ogilvy. “Delivering on the promise of AI search has traditionally required considerable in-house expertise and engineering resources to operate effectively. In addition to providing better search experiences, it also needs to be done reliably, quickly, and cost-effectively. Algolia has been the world leader in highly redundant, globally distributed instantaneous search using over 100 data centers worldwide. This global search distribution network combined with vector semantic search using extremely fast and efficient neural hashing technology is an exciting and truly unique solution.

Related articles:

Are neural networks the next big thing in research? launches new platform to overhaul e-commerce site search

Algolia launches additional AI models to boost user engagement for businesses

Amanda J. Marsh