Record Revenue, Massive Layoffs: The New Tech Paradox
A growing wave of tech industry layoffs is defying old logic. Companies are not shrinking because they are failing; many are cutting deeply while posting record results. Cloudflare is the latest flashpoint, announcing that it will lay off more than 1,100 employees—around 20% of its workforce—after a quarter where revenue grew 34% year-on-year to nearly USD 640 million (approx. RM2.95 billion). Management insists this is not classic belt-tightening. Instead, leaders describe AI as a powerful tailwind that is transforming how work gets done and what roles are still essential. Similar narratives are emerging across the sector, where strong financials coexist with restructuring announcements labelled as strategic pivots. This paradox is reshaping how workers interpret job security in an era where AI adoption, not revenue health, increasingly dictates which roles survive.
Cloudflare’s AI Push: Trimming Roles ‘Not Needed for the Future’
Cloudflare’s leadership frames its layoffs as a deliberate redesign for what it calls the “agentic AI era.” Internal data shows AI usage inside the company has surged more than 600% in just three months, with staff in engineering, HR, finance and marketing running thousands of AI agent sessions daily. CEO Matthew Prince argues that AI is driving a fundamental re-platforming of the internet and that many support roles no longer fit the company’s future. Engineering work is now heavily augmented by AI, including autonomous agents performing code reviews. Administrative support functions have automated large volumes of routine tasks, reducing the need for traditional back-office headcount. At the same time, Cloudflare plans to hire professionals skilled in leveraging AI and previously announced plans to recruit more than 1,000 interns to ramp up AI use, signalling a shift in favour of AI-powered teams rather than a smaller company overall.
GitLab’s ‘Different Kind of Layoff’ and a Shrinking Global Footprint
GitLab is also restructuring its workforce as it pivots toward becoming “the trusted enterprise platform for software creation in the AI era.” CEO Bill Staples stresses that its layoff process is not an AI optimization or pure cost-cutting exercise, even as AI clearly shapes the company’s direction. Instead, GitLab intends to reinvest savings into foundational AI infrastructure: agent-specific APIs, revamped CI/CD pipelines, richer data models, governance, and support for human-owned, agent-assisted and fully autonomous workloads. The company is opening a voluntary separation window and asking managers to hold detailed conversations with staff about their fit in the new AI-focused strategy. GitLab is also reevaluating its operational footprint, planning to reduce by up to 30% the number of countries where it has smaller teams. Together, these moves flatten management layers and redirect resources toward AI-centric capabilities, rather than sustaining legacy structures that may slow transformation.
From Traditional Roles to AI Specialists: How Workforce Restructuring Is Evolving
Across the tech industry, the pattern is becoming clear: companies are shedding roles considered “not AI enough” while hiring AI specialists and early-career talent comfortable with automation. Cloudflare’s restructuring leaves salespeople with direct revenue targets largely untouched but hits support, administrative and traditional engineering roles that can be augmented or replaced by AI agents. Other firms, from software platforms to fintech and social media players, have made substantial cuts in recent years, often citing changing workplace technology rather than collapsing demand. Workforce restructuring now centres on building AI-powered teams capable of leveraging tools that dramatically boost productivity. That means fewer layers of management, a smaller global footprint and more emphasis on roles that design, govern and orchestrate AI systems. For workers, the message is stark: being technically competent is no longer enough—future job security increasingly hinges on how integral one’s skills are to an AI-first operating model.