.png){kind=logo}
AI beyond coding: GitLab is deploying agents across planning, coding, testing, security, pipelines and releases.
Orbit creates context: Its software context graph could give agents the information needed to work faster and make fewer errors.
Enterprise opportunity: Growing revenue, large deals and reported ROI suggest GitLab’s broader AI platform strategy has significant commercial potential.
GitLab now faces a much bigger question than how well it can manage code and software delivery. Artificial intelligence has changed the value of the development platform itself. GitLab wants AI agents to handle work across plans, code, security, tests, pipelines and release control. That strategy could turn GitLab from a DevOps tool into the control layer for AI-driven software work.
GitLab made Duo Agent Platform generally available in January 2026. The platform lets teams use agents across the software lifecycle rather than restrict AI to code suggestions. GitLab says only about 20% of a developer’s time goes to code work, so faster code alone cannot create a similar rise in total delivery speed. The larger prize sits in the work around code.
GitLab 19.2 pushed that idea further. Duo CLI reached general availability and brought agents into the terminal. Custom Flows also reached general availability and let teams set repeatable agent workflows around GitLab events.
Security Review Flow added AI review for weak authorization, information leaks, business logic errors and race conditions. GitLab also added automatic fixes for vulnerable dependencies.
The August 20 release made the strategy more concrete. Flow Creator Agent lets a person describe an automation in plain English and receive a complete flow that can run through GitLab.
GitLab also added Secrets Manager as a paid add-on through GitLab Credits, plus bulk SAST false-positive detection and Agentic SAST Vulnerability Resolution. Security teams can select large groups of findings and ask GitLab to sort false positives and prepare fixes for real risks.
GitLab Orbit could provide the deeper advantage. Orbit creates a context graph that connects code, merge requests, pipelines, deployments, vulnerabilities and ownership. That gives an AI agent a view of the software system rather than a narrow view of source files.
GitLab says internal tests showed agents with Orbit responded up to 11 times faster, used up to 4.5 times fewer tokens and produced up to 45% fewer hallucinations. Orbit also offers open interfaces, so outside agents can use the same context.
Also Read - Loop Engineering: The Complete Guide to Building Smarter AI Agent Workflows
GitLab reported USD 286.3 million in revenue for the second quarter of fiscal 2027, up 21% from a year earlier. Non-GAAP income from operations reached USD 42.6 million, while the non-GAAP margin came to 15%. Net ARR rose 42% year over year, and gross bookings reached a company record. First orders rose more than 100%, while deals worth at least USD 500,000 rose more than 150%. GitLab also raised its full-year revenue outlook to USD 1.129 billion to USD 1.133 billion.
The results give weight to the wider platform strategy. Large enterprise deals suggest that customers can expand their use of GitLab as more software work shifts onto one system. The AI push has a clear commercial test: more agents, more usage, larger contracts and stronger customer retention.
A July Forrester study, commissioned by GitLab, found a 400% return on investment for a modeled organization that used Duo Agent Platform. The study showed a USD 7.5 million net present value over three years and a payback period of less than six months.
It also found 80% faster new developer setup, 75% faster code migration, 40% less time for quality assurance and security remediation, and a 20% gain in individual developer productivity.
The study covered four customer interviews and then built a composite company with USD 3 billion in annual revenue and 3,000 employees. The figures offer useful evidence, but they do not prove that every GitLab customer will get the same result. Forrester conducted the study, while GitLab sponsored the work.
Also Read: Google Flow AI: 7 Creative Tasks You Can Now Automate with Generative AI
The strongest part of the strategy comes from control, not another code assistant. AI can produce more code at a far higher speed, but enterprises still need context, security checks, approvals, audit records and clear ownership. GitLab already owns much of that workflow.
That gives the company a distinct path in the AI software market. GitHub, standalone code agents and model providers can help produce code. GitLab wants to manage what happens around that code. If agents become a normal part of software work, the platform that controls context, policy, security and delivery could hold more value than the tool that only writes the next block of code.
1. What is GitLab’s AI strategy
GitLab aims to use AI agents across the entire software development lifecycle, not just for code generation.
2. What is GitLab Orbit?
Orbit is a context graph connecting code, merge requests, pipelines, deployments, vulnerabilities and ownership to help AI agents understand software systems more deeply.
3. How is GitLab different from a coding AI assistant?
GitLab focuses on controlling the broader software workflow, including security, testing, approvals, governance and deployment.
4. What evidence supports GitLab’s AI opportunity?
GitLab reported strong revenue and enterprise growth, while a GitLab-sponsored Forrester study modeled a 400% ROI for Duo Agent Platform.
5. Why does AI make the DevOps platform more valuable?
As AI generates code faster, organizations need stronger context, security, testing, governance and delivery controls—areas GitLab already manages.
