Imagine it or not, internet search is nonetheless thriving as an business. As companies put money into utilizing AI brokers to make the most of their knowledge, there’s demand for instruments that not solely scrape the internet to inform what these AI bots do, but in addition return these ends in a manner that’s simpler to use with fashionable knowledge instruments.
That’s the promise behind internet search startup Nimble Way, which not too long ago raised a $47 million Collection B spherical, led by Norwest. The New York firm’s platform employs AI brokers to search the internet in actual time, verify, and validate the outcomes, after which construction the information into neat tables that may then be queried like a database.
That final half is essential right here. LLMs and AI brokers are nice for looking the internet, connecting outcomes from quite a lot of sources, and analyzing them, however they typically return the ends in plain textual content, which could be troublesome to work with at an enterprise stage. And that’s before you consider hallucinations, the threat of the agent misunderstanding your directions, or the use of unreliable sources.
By validating and structuring outcomes into tables, Nimble Method lets corporations use internet knowledge as if it had been already a part of their present databases. The startup additionally integrates with enterprise knowledge warehouses and knowledge lakes — massive centralized repositories the place companies retailer and analyze knowledge — provided by the likes of Databricks and Snowflake. Meaning its AI brokers can plug right into a enterprise’s trove of knowledge, utilizing it to construct context, and form how search outcomes are structured and returned.
In impact, this lets enterprises have stay, structured internet knowledge as a part of their present knowledge environments, Nimble Method CEO and co-founder Uri Knorovich (pictured above, center) informed TechCrunch.
Such integrations additionally enable Nimble Method’s software program to keep in mind constraints — similar to the way you need the search to be carried out, or which knowledge sources to faucet. This is significantly helpful for purposes similar to competitor evaluation, pricing analysis, know-your-customer (KYC) processes, model monitoring, deep analysis, and monetary evaluation. (Knorovich famous that Nimble Method works to guarantee all buyer knowledge stays inside clients’ knowledge infrastructure to adjust to knowledge retention and safety insurance policies.)
To that finish, the startup has partnered with Databricks, Snowflake, AWS and Microsoft to assist streamline enterprise deployments that require entry to inside knowledge sources. (Databricks additionally participated on this Collection B.)
Techcrunch occasion
Boston, MA
|
June 9, 2026
“Fashions can do loads of issues, however most manufacturing AI fails aren’t as a result of the fashions are not ok — it’s due to an information failure,” Knorovich stated. “What we’re seeing in the present day is that enterprises don’t want extra AI; they want AI with good, dependable internet search […] In the event you nail it down, when you can select what your agent can search and can’t search, this is the tipping level for enterprises to say, ‘hey we will really belief AI. We will really put AI to work in additional use circumstances’.”
Knorovich says the means to search the internet in actual time at scale, and validate and construction search outcomes, is what units Nimble Method aside from different knowledge brokers already in the area.
The startup presently has greater than 100 clients, with the majority of its income coming from massive enterprises, Fortune 500 corporations, and even some Fortune 10 corporations, together with main retailers, hedge funds, banks, and client packaged items corporations, in addition to some AI-native startups.
“Nimble is tackling an issue that has existed for years with out a correct answer and is now changing into of vital urgency,” Assaf Harel, associate at Norwest, stated in a press release. “Trusted stay internet knowledge is more and more changing into a prerequisite for AI brokers performing vital enterprise choices.”
The Collection B additionally noticed participation from returning traders Goal World, Sq. Peg, Hetz Ventures, Gradual Ventures, R-Squared Ventures, J-Ventures, and InvestInData. Proceeds from the spherical will likely be used to increase R&D in multi-agent internet search and a ruled knowledge layer that processes and validates search outcomes.
Nimble Method has now raised a complete of $75 million.
Disclaimer: This article is sourced from external platforms. OverBeta has not independently verified the information. Readers are advised to verify details before relying on them.
