---
title: When AI Delivers More, What Actually Counts | Quantum Sky
description: "Quantum Sky Chief AI Officer Cutter Brenton on moving beyond AI pilots and measuring what matters: mission outcomes, not lines of code."
image: https://engage.quantumsky.com/hubfs/QS-When-AI-Delivers-Featured-Image-v01.png
---

[![Quantum Sky\_Logos-WHITE](https://engage.quantumsky.com/hs-fs/hubfs/Quantum%20Sky_Logos-WHITE.png?width=6675&height=1875&name=Quantum%20Sky_Logos-WHITE.png "Quantum Sky_Logos-WHITE")](https://www.quantumsky.com/)

- [Home](https://www.quantumsky.com/)
- About 
    - [Capabilities](https://www.quantumsky.com/#capabilities)
    - [Project 137](https://www.quantumsky.com/#project-137)
    - [Careers](https://www.quantumsky.com/careers)
- [Contact](https://www.quantumsky.com/contact)

[![Contact Us](https://no-cache.hubspot.com/cta/default/2491502/bf2e78ec-5cbb-41b8-8432-a8dbeae62287.png)](https://cta-redirect.hubspot.com/cta/redirect/2491502/bf2e78ec-5cbb-41b8-8432-a8dbeae62287)

![QS-Blog-Banner-v02](https://engage.quantumsky.com/hs-fs/hubfs/QS-Blog-Banner-v02.png?width=2000&height=571&name=QS-Blog-Banner-v02.png "QS-Blog-Banner-v02")

# Quantum Sky Insights

## [When AI Delivers More, What Actually Counts](https://engage.quantumsky.com/blog/when-ai-delivers-more-what-actually-counts)

 Nov 12, 2026

- [Tweet](https://twitter.com/share)

*Quantum Sky Chief AI Officer Cutter Brenton joined industry leaders at the SpaceX AI executive breakfast to discuss how organizations can move beyond AI experimentation and turn emerging technologies into measurable mission impact.*

A Quantum Sky development team recently produced 246,000 lines of code over a 46-day project, demonstrating the speed at which AI is transforming software development. For Cutter Brenton, Quantum Sky’s chief AI officer, the accomplishment illustrated both the potential of AI-assisted engineering and the need to rethink how organizations measure success. The team’s output was impressive. But as Brenton reflected on the numbers, he raised a broader question: “Is that good?”

His question went beyond the volume of code produced. As AI accelerates development, traditional productivity metrics such as lines of code and commits tell leaders less than they used to. Every enterprise adopting AI now faces a larger question: How can organizations turn rapid technical progress into measurable mission impact? The answer depends on the capability delivered, the problem it solves and the outcome it achieves.

Brenton explored that challenge Oct. 29 during “Building the Super Intelligent Organization,” an executive breakfast at the Tower Club in Tysons, Virginia. He joined Carrie Lee, director of technology, federal health, at Guidehouse, and Chad Tetreault, field CTO at SpaceX AI, for a wide-ranging discussion on how organizations can move beyond AI experimentation and build lasting operational capabilities.

## Enterprise AI adoption: From pilots to measurable outcomes

For organizations investing in AI, the challenge is no longer simply demonstrating what the technology can do, but ensuring those capabilities translate into meaningful operational improvements. Lee offered an example of the gap between technical performance and organizational results, drawing on her experience with AI-assisted benefits processing at the Department of Veterans Affairs. VA used AI to automate parts of the process, with a human adjudicator reviewing each result. The adjudicators were still rated on accuracy, and nobody changed that. “They didn’t trust anything the AI did,” Lee said. “So they’d go back to every step of the process and redo it.”

Instead of removing manual work, the implementation preserved much of it, and the efficiency gains never arrived. Lee said too many organizations stop at the demonstration. “Everyone wants to pilot,” she said. “But then, once it’s out of pilot, how do you actually get adoption?”

Her answer is to redesign the work around the technology rather than bolt the technology onto the work. The challenge of enterprise AI adoption is therefore as much organizational as technical. For federal leaders, performance measures, review requirements and accountability belong in the AI adoption strategy from the start, rather than as cleanup after a stalled pilot.

The adoption challenge Lee described also raises a question for executives: What result was the AI initiative meant to deliver in the first place? Brenton noted that much of today’s AI activity starts from the bottom up. Someone automates a task, the result travels up the chain and the board eventually asks how much money it saved, a return nobody designed the project to deliver.

The fix is to settle the purpose first. “What outcomes are we trying to achieve?” Brenton asked. Increasing efficiency and cutting costs call for different designs. An agency that needs more processing capacity shouldn’t be judged against a cost-cutting target, and a security operations center that needs faster detection shouldn’t be judged on headcount.

Brenton also rejects the assumption that AI’s value has to come from cuts. “You can design initiatives with paired agent and human workflows that do impact the bottom line in a positive way without removing people from the organization,” he said. Whether to invest in partnering people with agents or to pursue cost reduction, he argued, is a choice executives should make deliberately rather than by default.

## Building AI capabilities that scale

At Quantum Sky, that emphasis on measurable outcomes is reflected in Project 137, the company’s roughly 50-person research and development organization. Brenton described an engineering model that brings technical teams closer to the missions, operational challenges and customers their solutions are designed to support.

Through interdisciplinary squads, engineers collaborate with growth teams and subject matter experts to shape requirements, develop capabilities and take greater ownership of the results. The approach reflects a broader shift in AI-enabled engineering: Success is measured not simply by how quickly a team can build something, but by how effectively that capability addresses a real operational need. Brenton described the attitude he wants from his engineers: “Look what I built. Don’t you want to use this?”

Quantum Sky is also applying that approach to enterprise AI agents, building on personalized assistants, workflows and automations that employees are developing to support their daily responsibilities. Their work reflects the rapid adoption of agentic capabilities and the opportunity to turn individual innovations into shared enterprise resources.

Doing that means creating shared capabilities that can be reused, governed and scaled across teams. Enterprise AI governance requires consistent standards, appropriate data access and flexibility to support different mission requirements. For Brenton, the challenge is not simply putting more agents to work, but building an environment where they can deliver value across the organization.

Cybersecurity offers an example of how agentic workflows can operate at scale. Brenton described agents automating tier-one security operations tasks and conducting threat hunts under human direction. “You’re actually combing the network with a digital workforce,” he said.

Brenton expects the next wave at the tactical edge, where sensors are multiplying and bandwidth can’t be guaranteed. Small language models running close to the sensor would let agents work where the data is collected. “Instead of moving data around through pipes, you’re going to be moving inference and moving workflows through pipes,” he said.

That enterprise approach also depends on making the right data available to the right capabilities. Lee emphasized the importance of providing agents with the information and organizational context they need to operate effectively. Without that foundation, she warned, companies risk developing isolated AI capabilities that cannot easily work together.

## Why mission expertise matters more than ever

The ability to scale these systems makes mission expertise more important, not less. As AI lowers the cost of writing code, Brenton said the expertise Quantum Sky prioritizes is changing. “I’m trying to buy context. I’m trying to buy mission expertise,” he said. In defense, intelligence and other regulated settings, that context is often technical in its own right: understanding how sensors work and how a process flows can matter more than knowing how to code.

It also changes what engineers own. “We’ve made it so easy to code that now we have to think about, ‘What do we take ownership back of?’” Brenton said. “And it’s the architecture.” People keep responsibility for design, scale, risk and rollout, while agents increasingly handle the build.

Lee sees the same shift spreading beyond IT. On one team she works with, analysts who review medical claims for fraud, waste and abuse used AI agents to build their own detection rule sets, work that once required data engineers. “Every individual in the organization is becoming a technologist through the use of AI,” she said. The people with the context are becoming the builders.

That shift is moving faster than most planning cycles. No one on the panel claimed to know where AI will be in two years, an awkward fact for agencies that request funding two to three years before it arrives. Brenton compared the moment to an ultramarathon over rolling hills, unlike a cloud migration with a visible finish line. “You don’t know what’s on the other side of the hill, but you know that we’re going to crest some of these hills,” he said.

Lee’s answer for government is to budget based on capabilities rather than on specific technologies. At VA, that meant moving to product-based budgeting, which let leaders shift money between people and AI as needs changed.

As AI agent orchestration becomes part of everyday operations, organizations will need employees who can coordinate multiple systems while maintaining oversight of their outputs and decisions. Brenton’s near-term forecast is clear. “If you’re not planning in the next year to have all of your staff, regardless of technical competency, orchestrating at least five or six agents as part of their day-to-day routine, you’re crazy,” he said. “You have to prepare for it today.”

AI will keep making output cheaper. Lines of code, commits, agents and pilots will keep piling up on dashboards. The organizations that pull ahead will be the ones that decide what success means before the numbers arrive. Then, when someone looks at the numbers and asks Brenton’s question, “Is that good?” they’ll already have an answer.

***About Quantum Sky:** Quantum Sky partners with government and defense organizations to prepare for emerging technologies. As quantum computing moves from possibility to reality, the company delivers cybersecurity, networking, software, AI, and post-quantum solutions that help customers modernize critical systems, strengthen resilience, and maintain a strategic advantage. Headquartered in Reston, Virginia, Quantum Sky has 11 offices across the United States and Europe.*

![Cutter Brenton](https://engage.quantumsky.com/hs-fs/hubfs/Blog/QS-When-AI-Delivers-Headshot-Cutter-v01.png?width=100&height=100&name=QS-When-AI-Delivers-Headshot-Cutter-v01.png)

#### Written by [Cutter Brenton](https://engage.quantumsky.com/blog/author/cutter-brenton)

Cutter Brenton is Chief AI Officer at Quantum Sky, where he leads the company’s enterprise AI and quantum strategy, including Project 137.

### Post a Comment

### Subscribe to Our Blog

- Recent
- Popular
- Categories

### Lists by Topic

### Posts by Topic

#### ABOUT US

The world the mission operates in is going post-quantum, contested, and machine-speed. Quantum Sky engineers the advantage across cyber, networks, software, and quantum because the mission demands dominance, not parity. We don't follow the map. We draw it.

[Learn More](https://www.quantumsky.com/)

#### USEFUL LINKS

- [Home](https://www.quantumsky.com/)
- [Company Overview](https://www.quantumsky.com/#capabilities)
- [Leadership](https://www.quantumsky.com/#leadership)
- [Project 137](https://www.quantumsky.com/#project-137)
- [Contact Us](https://www.quantumsky.com/contact)

#### CONTACT US

[703-885-7900](tel:+17038857900) 

[bd@quantumsky.com](mailto:bd@quantumsky.com)

11950 Democracy Dr, Ste 650   
Reston, VA 20190-5692

<https://www.linkedin.com/company/gotyto/>   <https://x.com/WeAreQuantumSky>   <https://www.facebook.com/WeAreQuantumSky/>

 Copyright 2026 Quantum Sky Engineering LLC. All rights reserved.

\[fa icon="chevron-up"\]

[![](https://engage.quantumsky.com/hubfs/Quantum%20Sky_Logos-FullColor.svg)](https://www.quantumsky.com/)

- [Home](https://www.quantumsky.com/)
- About 
    - [Capabilities](https://www.quantumsky.com/#capabilities)
    - [Project 137](https://www.quantumsky.com/#project-137)
    - [Careers](https://www.quantumsky.com/careers)
- [Contact](https://www.quantumsky.com/contact)

[![Contact Us](https://no-cache.hubspot.com/cta/default/2491502/f6335e0a-ca80-4263-9ce6-0ce3da5ffda7.png)](https://cta-redirect.hubspot.com/cta/redirect/2491502/f6335e0a-ca80-4263-9ce6-0ce3da5ffda7)

```json
{
  "@context" : "https://schema.org",
  "@type" : "BlogPosting",
  "author" : {
    "@type" : "Person",
    "name" : "Cutter Brenton",
    "url" : "https://engage.quantumsky.com/blog/author/cutter-brenton"
  },
  "dateModified" : "2026-11-12T21:38:41.419Z",
  "datePublished" : "2026-11-12T21:38:41.000Z",
  "headline" : "When AI Delivers More, What Actually Counts | Quantum Sky",
  "image" : [ "https://engage.quantumsky.com/hubfs/QS-When-AI-Delivers-Featured-Image-v01.png" ],
  "mainEntityOfPage" : {
    "@id" : "https://engage.quantumsky.com/blog/when-ai-delivers-more-what-actually-counts",
    "@type" : "WebPage"
  },
  "publisher" : {
    "@type" : "Organization",
    "logo" : {
      "@type" : "ImageObject",
      "url" : "https://engage.quantumsky.com/hubfs/Quantum%20Sky_Logos-FullColor.svg"
    },
    "name" : "Quantum Sky"
  }
}
```