For years, software teams measured engineering output with familiar metrics such as lines of code, commit counts, and deployment frequency. AI coding assistants have changed that equation. A single prompt can now generate thousands of lines of code in seconds, making old productivity measurements far less reliable. That shift has created a new problem for engineering leaders and CFOs alike: AI spending is climbing, yet few companies can clearly explain what they are getting in return.
San Francisco-based Weave believes it has the answer. The engineering intelligence startup announced Tuesday that it has raised a $13.5 million Series A funding round to help companies measure the real business impact of AI-assisted software development and eliminate what it calls “tokenmaxxing,” a growing practice of optimizing AI coding tools for token volume rather than meaningful engineering progress.
The round was led by Standard Capital, with participation from Y Combinator, Moonfire, Burst Capital, IrregEx, and the Agent Fund. The fresh capital will support product development and expand Weave’s platform as enterprises spend more on AI coding tools and look for better ways to measure their return on investment, the company said in a blog post.
The announcement comes at a time when AI coding assistants have become standard across many engineering organizations. Businesses are investing heavily in tools from OpenAI, Anthropic, Google, GitHub, Cursor, and others, yet finance executives often struggle to connect soaring AI bills with measurable business outcomes. That gap has sparked a broader conversation about whether existing engineering metrics still reflect real productivity in an era where AI generates a growing share of software code.
With $13.5M in funding, Weave aims to stop ‘tokenmaxxing’ and measure AI engineering output
Weave’s pitch is straightforward. Rather than counting lines of code or commits, its platform analyzes both human and AI contributions across the software development lifecycle, measuring engineering output through machine learning models that attempt to connect coding activity with actual business progress.
According to the startup, its platform has already analyzed more than 2 million human and AI code contributions across over 20,000 engineers working at more than 500 organizations. Customers include Robinhood, Reducto, and PostHog.
The company’s software tracks pull requests, code reviews, deployments, and AI-generated contributions before combining that information into a single measurement system. It gives engineering executives visibility into team productivity, AI adoption, engineering quality, and the cost of AI-generated work. One feature estimates how much AI spending is required to produce an hour of completed engineering work. Another recommends lower-cost AI models based on production data without sacrificing quality or development speed.
“Our goal is to ship the highest quality product as quickly and efficiently as possible for our customers. We use Weave to get objective measurement of our teams and agents as well as ways to optimize,” said David Casem, Co-Founder & CEO at Telnyx.
At the center of Weave’s strategy is its criticism of “tokenmaxxing.” The term refers to the tendency for developers or AI systems to maximize token usage simply to generate more code, even when that output does little to improve a product. The company argues that traditional engineering metrics unintentionally reward this behavior by treating larger volumes of code as a sign of higher productivity.
“The era of tokenmaxxing is over”, said Adam Cohen, Founder and CEO of Weave. “Every engineering and finance leader is now asking the same question: what is our AI spend actually returning? Lines of code are dead as a metric. Weave gives leaders one objective measure of real output, human and AI, so they can finally manage engineering like every other part of the business.”
Investors see the issue as becoming more significant as AI spending moves from experimental budgets into core business operations.
“AI spend is the most powerful force in the world, and right now there is not an easy way to measure it” said Dalton Caldwell, General Partner at Standard Capital. “The opportunity for Weave is to enable every organization to effectively track and route their spend, and is thus is a critical piece of infrastructure for any organization embracing AI.”
Weave enters a market that has become increasingly competitive as companies search for better ways to evaluate developer productivity in the AI era. Engineering analytics platforms have traditionally focused on delivery metrics such as deployment frequency and development cycle time. AI-assisted coding has introduced a new variable by blurring the line between code generation and meaningful engineering output.
That shift has created demand for tools that measure outcomes instead of activity. Business leaders are asking a different question than they did a few years ago. The goal is no longer to determine whether engineers are writing more code. It is to determine whether AI spending leads to faster product delivery, better software quality, and stronger business results.
With fresh funding, Weave is betting that engineering intelligence will become a standard part of enterprise AI adoption. If companies continue pouring billions into AI coding tools, investors and executives alike will expect clearer answers about where that money is going and what value it creates. For startups building software infrastructure around AI, that question could become one of the biggest opportunities in enterprise technology over the next few years.
