Neurometric Raises $4M in Funding by Mu Ventures

The round saw the Backers include Betaworks, ex-Ante, and Everywhere.VC, Encoded, Vermillion, Abstraction, and Mu Ventures and angels, including Jason Calacanis, co-host of the All-In Podcast, and Dharmesh Shah, CTO of Hubspot.

The company plans to use the funding to grow its engineering and AI research teams and add more optimization tools to its core platform.

As companies move AI agents into production, a single workflow can involve dozens of AI model calls. Many businesses still use expensive frontier models for every task, even when smaller, lower-cost models can deliver the same results.

Read More:Arca Raises $48.5M in Series A Funding Led by General Catalyst

Neurometric analyzes each AI request, improves prompts when needed, and routes every task to the most cost-effective model that meets the required performance. If no existing model is suitable, it creates a custom small language model. For high-volume, simple tasks, the platform automatically builds specialized small models to improve speed and reduce costs.

Many companies still rely on manual testing and separate tools to decide which AI model should handle each task. These decisions can quickly become outdated as new models are released and prices, speed, and performance change.

Neurometric brings model routing, custom small language model (SLM) creation, and a marketplace of pre-trained SLMs into one platform. Its Task Endpoint Manager automatically selects the best model for each request based on accuracy, cost, and speed. If no existing model is suitable, the platform creates a custom SLM for that task. Customers can also use ready-made SLMs for common workloads through its marketplace.

This approach lets companies use powerful frontier models only for tasks that require them, while handling simpler tasks with smaller, lower-cost models.

In early customer deployments, Neurometric’s routed and custom-built models have delivered accuracy improvements of up to 20 percentage points over frontier models, while also reducing costs and improving response times.

"Frontier intelligence will keep getting less expensive, but companies will also consume far more of it," May said. "The winners will not be the businesses that simply buy the most tokens. They will be the ones that know where advanced intelligence creates value and where a smaller model can do the job just as well."

Alex Benik from Encoded, and investor in the round, stated "Neurometric is tackling one of the most pressing problems in the AI ecosystem today. The team has a unique mix of AI talent and systems engineering experience that positions them well to take on this task and help companies optimize their token spend at multiple layers of their infrastructure."

"Companies need to know where frontier-level performance is worth paying for and where a smaller model can deliver the same result at a fraction of the cost. That discipline will determine whether agentic AI can move from promising pilots to a business model that scales," said Neurometric COO Calvin Cooper.

"The number of available models is growing too quickly for companies to evaluate every option by hand," May said. "And the number of tools and techniques to improve them is growing even faster. Things change so fast that a human token engineer can't keep up. That decision needs to be automated and continuously reevaluated as the market changes."

"Companies have spent the past year proving that AI agents can perform increasingly complex work. Now they have to prove the economics still make sense when those agents are operating at scale," said Rob May, CEO of Neurometric. "Every model call is also a pricing decision, and those decisions compound across an agent's workflow. Token engineering gives companies a way to control that cost without sacrificing quality."

About Neurometric

Founded by Rob May, Byron Galbraith, Calvin Cooper, and Dave Rauchwerk, Neurometric is an AI optimization platform. It automatically routes each AI task to the most cost-effective model that meets the required quality and creates custom small language models when existing models are not the right fit.

Read More:Upside Raises $20M in Series A Funding Led by Aquiline