Nimble launches Web Search Agents to cut AI research token costs
Web search platform company Nimble today launched Web Search Agents, a product that learns a customer’s domain and then runs complex web research tasks on its own.
The company is aiming the release at teams that have found general-purpose web search too blunt for production agents. Generic tools return a wide, unstructured set of results and leave the agent to work out what is actually relevant, an approach that burns tokens on unnecessary tool calls and on processing pages that do not matter.
Nimble’s pitch is that a market research agent and a lead enrichment agent should not get the same product. Its harness self-learns the knowledge work involved in a task and adapts retrieval strategies to it, combining proprietary indexes with real-time retrieval from live sites to pull what the company describes as the freshest and most granular information available.
Benchmark testing published by Nimble showed a 21-point increase in answer quality and 51% fewer tokens spent per query.
“For enterprises, the real bottleneck is accuracy and cost,” said Uri Knorovich, chief executive and co-founder of Nimble. “Agents need the right live web context without wasting tokens or relying on generic search results.”
Speed is not the pitch. Nimble is going after agents that run for hours on business-critical research, where a missed source costs more than a slow answer. The harness checks its own work as it goes, which the company says leaves less for the team behind it to redo.
Rox, an AI-native customer relationship management company, reported a 20-fold reduction in token costs after adopting the service, alongside better quality and completeness in the information reaching its agents. Almog Lavi, head of product at code integrity startup Qodo, said in the announcement that Nimble let his team tune a Claude Managed Agent to surface the competitor signals that mattered to them rather than a generic market summary.
Web Search Agents is available through Nimble’s application programming interface, its software development kit and Model Context Protocol, with a free trial. Developers can wire it in as a tool inside an existing agent or build applications on top of it, from low-latency search through to deep research and structured dataset generation. Nimble says the platform is fielding more than 90 million searches a day. Its customer base takes in Fortune 500 enterprises as well as younger AI-native firms.
Founded in 2021, Nimble is a venture capital-backed company that most recently raised $47 million in a February Series B round led by Norwest Venture Partners LP. Databricks Ventures, Target Global Management GmbH, Square Peg Capital Pty. Ltd., Hetz Ventures, Slow Ventures, R-Squared Ventures, J-Ventures and InvestInData also participated. The New York company has raised $75 million to date.
Image: SiliconANGLE/Ideogram
Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.
15M+ viewers of theCUBE videos, powering conversations across AI, cloud, cybersecurity and more** 11.4k+ theCUBE alumni**— Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network.
About SiliconANGLE Media
theCUBE AIand theCUBE SuperStudios — with flagship locations in Silicon Valley and the New York Stock Exchange — SiliconANGLE Media operates at the intersection of media, technology and AI.
Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.