The software development space is currently undergoing a significant transformation. At SPK and Associates, we have spent over 20 years helping engineering organizations navigate the complexities of ALM, DevOps, and regulated product development. We have seen the shift from waterfall to agile, and from on-premise silos to integrated DevSecOps platforms. Today, we are standing at the threshold of the new era of agentic engineering. GitLab recently announced a major strategic pivot, internally referred to as “Act 2.” This is a “generational rebuild” of how software is conceived, built, and governed. For engineering leaders in regulated industries, understanding this shift is critical. It represents a move toward a future where software is built by machines but directed by people.
Why GitLab is Moving Toward Agentic Engineering
Technology should reduce friction for engineers, not create more administrative work. However, as AI capabilities have rapidly advanced, many platforms have struggled to keep pace. They often add on AI features to the previous infrastructure that was not designed for those types of workloads. GitLab recognized that this was not the way to operate. Their existing operational shape, while successful for the DevSecOps era, was not optimized for the speed and volume of the agentic era.
Historically, the cost and time required to produce and manage code were primary bottlenecks. With AI agents now capable of planning, coding, reviewing, and deploying, those bottlenecks are disappearing. This is why GitLab is making the shift from a more traditional DevSecOps platform to becoming a trusted enterprise platform for autonomous software creation. This shift is a proactive response to a new reality. A reality where the value of software for builders is increasing, but so is the amount of software and builders. GitLab had to rewire its internal processes to match the speed of the technology they are building.
What this Shift at GitLab Entails
The restructuring at GitLab is a deliberate alignment of their internal operations with their “Agentic Engineering” thesis. This involves several key changes designed to increase velocity and empower technical teams:
1. Flattening the Organization
GitLab is removing up to three layers of management in certain functions. The goal is to bring leadership closer to the actual engineering work. In complex engineering environments, layers of bureaucracy often lead to requirements drift and communication silos. By flattening the structure, GitLab aims to reduce the distance between strategic decisions and technical execution.
2. Empowered, End-to-End R&D Teams
The R&D organization is being reorganized into roughly 60 smaller, independent teams. These teams have end-to-end ownership of their domains. For clients, this means more frequent updates and a faster response to the specific needs of the market.
3. Internal “Customer Zero” Strategy
GitLab is rewiring its internal processes with AI agents. They are automating reviews, approvals, and handoffs. By becoming “customer zero” for their own agentic platform, they are identifying the friction points of autonomous workflows before they ever reach their clients.
4. Reengineering for Machine Scale
Perhaps the most significant technical change is the “generational rebuild” of their underlying infrastructure. Traditional Git-based systems were designed for human-rate commits. AI agents can push commits and trigger pipelines at a rate no human team could ever match. GitLab is moving from a monolith to modern, API-first, composable services designed to handle 100x the current scale.
How This Agentic Engineering Shift Impacts Clients
For engineering directors and VPs, particularly those in regulated environments, these changes offer several practical benefits. While structural news can sometimes cause concern, GitLab’s “Act 2” is designed to provide greater operational confidence and faster delivery.
Enhanced Traceability and Governance
In regulated industries, governance can often slow down operations. GitLab has always had security and governance at the core of its operations, but the new strategy ensures governance is not just an added layer. As agents take on more work, the platform automatically enforces policy, proves what happened and why, and keeps sensitive data secure. This “Governance at the Core” approach allows teams to move fast without sacrificing audit readiness or compliance with standards like ISO 13485 or ISO 26262.
Context-Driven AI
One of the biggest challenges with generic AI tools is the lack of context, which leads to hallucinations or irrelevant code. GitLab is investing in a connected data model that spans planning, code, security, and operations. This unique context allows AI agents to deliver better results with fewer tokens. For DevOps teams, this means AI suggestions that actually understand their specific architectural constraints and historical project data.
Orchestration Over Activity
Enterprises need running software that moves the business forward, which is why GitLab is reimagining CI/CD as an orchestration service. This service coordinates multiple agents, manages state, resolves conflicts, and keeps a human in the loop for important, high-judgment decisions. This reduces the administrative burden on your senior engineers, allowing them to focus on system design and complex problem-solving.
Flexible Business Models
As the way we build software changes, the way we pay for it must also evolve. This is why GitLab is introducing more flexibility by mixing traditional subscriptions with consumption-based pricing for the work agents perform. This allows engineering organizations to scale their costs in alignment with the actual value and volume of work being delivered.
The Agentic Engineering Shift at GitLab
The shift toward agentic engineering is not about replacing engineers, but elevating them. As GitLab notes, the supply of deep technical problems is multiplying, and the engineers who can solve them will be more valuable than ever. At SPK and Associates, we believe that technology should always serve the business goal. GitLab’s restructuring is a bold step toward ensuring its platform can handle the massive scale of the AI era while still providing the governance and context that engineering organizations require. If you are looking to modernize your software delivery process or want to understand how GitLab’s new agentic direction can benefit your specific regulatory environment, contact us today.
