TLDRocket
Sign in

NextLM brings its precision prospecting agent to Google Cloud Marketplace

SiliconANGLE Paul Gillin

NextLM put its sales-lead AI on Google Cloud Marketplace and Gemini Enterprise. It taps Google spend, but the real pitch is finding the person behind the company, not just the visitor.

Based on reporting by SiliconANGLE, Paul Gillin — read the original for the full story.

Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error

NextLM is moving its AI Prospecting Agent onto Google Cloud Marketplace and into Gemini Enterprise, giving sales teams a new way to buy and use the tool inside Google’s stack. The New York startup says the software is built to spot individual people who are likely researching a customer’s product, then show sellers why the system made that call.

That focus on individual intent is the whole point. B2B sellers usually know a company is browsing, but not who at that company is actually looking. NextLM tries to close that gap by combining behavioral signals from advertising platforms and public databases, then ranking people it thinks are closer to a purchase. The company says its engine processes 35 billion behavioral signals a day and covers more than 300 million companies and 370 million professional profiles.

Chief executive Chris Anzalone said the model is not chasing single clues in isolation. It looks for combinations of behavior that line up in ways that suggest buying intent. His example was simple: someone reading a real estate site might just be curious, but if that same person is also looking up nearby schools and restaurants, the odds look better. The agent only surfaces prospects in the top 10% of its scores, and each customer gets a private version of the model that learns from that customer’s own wins and losses without feeding deal data back into the shared base model.

The underlying model, called Savant, is built on Nvidia’s Nemotron architecture. NextLM says it trained Savant on Nvidia DGX Spark systems and runs it on Nvidia A100 GPUs hosted on Google Cloud. In a company-published benchmark, NextLM said Savant improved capture of eventual buyers among its highest-scored prospects by 2.45 times versus nine general-purpose AI models. It also said scoring costs run from 3 cents to 11 cents per thousand prospects on an owned GPU.

The privacy questions are obvious, and NextLM doesn’t pretend otherwise. Anzalone said the company can’t identify every person it sees in the data, and only names prospects when it believes there is a 97% likelihood to buy. The company says it has an internal compliance checklist, depends on data partners’ consent assurances, and excludes users covered by Europe’s GDPR. Pricing starts at $599 a month with no long-term contract, and customers draw down credits as they request prospects and research. Google’s reach just makes the product easier to buy; NextLM says it has no plans to move its infrastructure elsewhere.

My take — AI-written commentary, not fact-checked reporting

This is classic enterprise AI: not a chatbot, just a sharper sales machine with a nicer cloud checkout flow. The pitch is less “magic intelligence” than “we know who to bother,” which is exactly why it will sell. And yes, the privacy discomfort is the point — nobody builds a prospecting engine this precise because they enjoy restraint.

Read more about this at: SiliconANGLE

Related stories

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.