Understory ranks No. 140 on Inc. 5000 list of fastest-growing companies in U.S. Read more about Understory on the 2026 Inc. 5000 list

Bill Staples

Bill Staples

Chief Executive Officer at GitLab

United States · Developer tools · LinkedIn profile

Bill Staples is Chief Executive Officer at GitLab, based in United States. On LinkedIn Bill writes mostly about company news and industry analysis, most often as text posts.

22K+followers
104reactions on a typical post
228reactions on the best post
5 of 30recent posts worth featuring

Bill's best LinkedIn posts

Recent posts that drew 100 reactions or more and say something a peer would learn from, most reactions first. Click any post to open it on LinkedIn.

Bill Staples
Bill StaplesChief Executive Officer at GitLab
Every CEO is asking some version of the same question: We’re spending more on AI. Where is the return? Tokens consumed, code generated, and developer activity don’t answer that question. The unit of value in software engineering is a change that makes it safely into production. That’s why I believe one of the most important metrics for the next era of software engineering will be cost per accepted change. How much does it actually cost to move a change from intent through implementation, review, testing, security, and acceptance? And is AI bringing that number down? At Transcend, we introduced early access to GitLab Duo Agent Platform Impact Analytics, connecting AI consumption and workflow results by team, task, and model. It’s an important step toward measuring outcomes instead of activity. The winners in AI will be the ones that turn intelligence into trusted outcomes most efficiently.
Bill Staples
Bill StaplesChief Executive Officer at GitLab
The economics of AI are changing as fast as the models. For the last few years, much of the conversation has focused on which model is “best.” With open weight models gaining momentum, I think the objective becomes: use the least expensive model that can reliably produce an accepted result for the task. Today GitLab expanded the hosted open-weight models available in Duo Agent Platform. Customers can now choose models optimized for different combinations of quality, latency and cost. The price/performance differences are meaningful. Some of these models provide up to 4x more calls per GitLab Credit, while Kimi K3 and GLM 5.3 performed strongly against comparable frontier models in our internal testing. That matters a lot more in an agentic world. A single user request can result in multiple model calls, and long-running agents may repeat that process over and over. Small differences in inference economics compound quickly at scale. So model selection increasingly becomes an optimization problem: apply expensive intelligence where it creates incremental value, and use more efficient models everywhere else. And do it while maintaining centralized security, governance and control over which models are available. Ultimately, I don’t think the metric that matters is cost per token or even cost per model call. It’s cost per accepted change. That’s the economic unit we need to keep driving down as software development becomes increasingly agentic. Read more in the blog linked in comments below.
Bill Staples
Bill StaplesChief Executive Officer at GitLab
I recently sat down with Darren Wilkes, CTO at COCC, to talk about how his team is using GitLab Duo Agent Platform. COCC powers core banking and digital banking for roughly 175 banks and credit unions managing more than $100 billion in assets. His team moved from manual regression testing and mismatched environments to a GitLab pipeline where developers work in an environment that mirrors production. The pipeline catches vulnerabilities before code ever ships. Darren's team built hundreds of agents in GitLab Duo Agent Platform that pick up requirements from product and carry them through the SDLC. Having a full history in GitLab makes every AI decision traceable, so when something needs fixing, the team can quickly get back to baseline. COCC is now using AI to test its own code and modernize legacy applications into microservices with GitLab Duo. Thanks, Darren, for the partnership and for sharing what your team is building.
Bill Staples
Bill StaplesChief Executive Officer at GitLab
When implementation becomes abundant, trust becomes scarce. I wrote that a few weeks ago in When Code Is Abundant. Security is where it gets concrete fastest. Dario Amodei’s essay on pacing the frontier reinforces the same concern: AI is compressing the time and cost required to attack, while too much of defense still operates at human speed. There’s another front line getting less attention: the software factory itself. Agents now read source, call tools, change infrastructure, open merge requests, and ship software, driven organically through shadow software factories that sit outside the organization's controls which govern how software reaches production. The security model has to change with them. Three changes I think every enterprise should make now: 1. Treat every agent as a privileged actor, with policy in the execution path. Identity, least privilege, bounded tools, defined parameters, traceable actions, and explicit approval for high-impact decisions. This is how you control what can run, what must be scanned, what requires approval, and what can ship should be enforced in the workflow and verified, not documented and hoped for. 2. Measure time from detection to verified remediation. If attackers can find and exploit weaknesses autonomously, remediation measured in days or weeks is an exposure. A scanner that grows the backlog faster than the team can drain it has not solved the problem. It has to be machine-scale remediation to match machine-scale detection. 3. Keep an eye on customer experience and security. The change rate increases as you quickly remediate vulnerabilities. Optimizing for security posture OR customer experience is a false choice. You can do both. Make sure your build pipeline has the customer experience verification gates by default. The answer is to govern the path into production with identity, policy, audit, customer experience verification, and remediation built in, while still letting teams use the best model for the job. The model will change. The control plane shouldn’t. As software creation becomes agentic, security can’t remain a collection of gates around the factory. Security has to become part of the factory. Machine-speed creation requires machine-speed verification and remediation. Our CISO, Chaim Mazal, will take this much further with GitLab’s full point of view: a living standard for securing the software factory in the agentic era.
Bill Staples
Bill StaplesChief Executive Officer at GitLab
I recently sat down with Victor Rodrigues, VP of Customer Software Strategy, Architecture, and Systems at Cochlear, to talk about building software for hearing implants. Cochlear makes implantable hearing devices and the external sound processors that go with them. Software runs inside both, alongside the tools surgeons and audiologists use and the software patients use to manage their devices. Cochlear has used GitLab for 14 years. What began with a small group grew into the company standard as more teams gravitated to it over other tools. Because the work in GitLab is codified, it stays traceable and auditable. AI is changing how Cochlear builds software. With GitLab Duo, setting up security scanning takes about 30 minutes instead of two weeks, and people who don't normally write code are orchestrating code through AI tooling, all of it managed in GitLab. To give that work more structure, Cochlear is testing GitLab Orbit, our knowledge graph, which maps a large codebase so AI has more context. Thanks, Victor, for the partnership and for sharing Cochlear's story.

What Bill posts about

  • Company news53%
  • Industry analysis13%
  • Customers10%
  • Advice and playbooks7%
  • Team and hiring7%
  • Leadership7%
  • Events and content3%

What makes the best ones work

  • Most of Bill's best posts open with a plain statement of a view.
  • 60% of them are text only, with no image or video.
  • They run long, often past a thousand characters.

Questions about Bill Staples

Who is Bill Staples?

Bill Staples is Chief Executive Officer at GitLab, based in United States. On LinkedIn Bill writes mostly about company news and industry analysis, most often as text posts.

What does Bill Staples post about on LinkedIn?

Mostly company news, industry analysis and customers. Of the 30 recent posts we read, 5 drew 100 reactions or more with real substance.

What is Bill Staples's most popular recent LinkedIn post?

The one that opens “Every CEO is asking some version of the same question:”, with 228 reactions and 51 comments.

Every post belongs to Bill Staples. Understory is not affiliated with Bill Staples or GitLab. We read public LinkedIn posts to study what B2B leaders write and what their audiences respond to, and the bio is our own summary. If this is you and you would like the page changed or removed, email alex@understoryagency.com and we will do it.

More developer tools thought leaders

See every developer tools thought leader, or the ads GitLab runs.

Want posts like these on your profile

We ghostwrite LinkedIn posts for B2B founders and executives, in their own voice, and turn the audience into pipeline.