Platform engineering, once a refinement of DevOps practices, now stands at a crossroads where artificial intelligence reshapes its core mission. What began as a drive to streamline developer workflows through internal platforms has expanded to encompass APIs, data pipelines, and AI agents, demanding unified governance across silos. This evolution promises to accelerate software delivery but raises thorny questions about ownership and control in enterprise environments.

At the heart of this shift is the recognition that isolated tools no longer suffice. As devm.io reports, modern platforms integrate infrastructure resources, developer tools, services, APIs, data capabilities, and AI models. ‘Platform Engineering didn’t suddenly become something else. It quietly became more than we initially planned for,’ notes Sebastian Meyen, head of content at Software & Support Media.

The upcoming Platform Engineering Week in London, set for May 11–15 at Park Plaza Victoria, underscores this convergence by blending DevOpsCon, API Conference, and MLcon. Organizers aim to mirror real-world intersections of these domains rather than siloed discussions.

Evolution from DevOps Foundations

DevOps broke down barriers between development and operations, but as organizations scaled, developers drowned in infrastructure complexity. Platform engineering emerged as the response, creating self-service internal developer platforms (IDPs) that abstract away toil. According to Roadie, Gartner forecasts 80% of large engineering organizations will have platform teams by 2026, up from 45% in 2022, providing reusable services for application delivery.

This maturation reflects broader adoption. The 2025 DORA report, cited by PlatformEngineering.org, reveals nearly 90% of enterprises now operate internal platforms, surpassing Gartner’s timeline. ‘We are not arguing as much about what should be within an engineering platform,’ says Ricky, a platform expert. Conversations have advanced to mature capabilities like product thinking and enterprise alignment.

Core elements include infrastructure automation, CI/CD management, security, observability, and API governance—12 foundational pillars that successful teams deliver as business enablers.

AI as the Convergence Catalyst

AI accelerates this trajectory, moving from coding assistants to agentic systems that demand robust platforms. The New Stack highlights AI’s role in amplifying productivity: At Spotify, AI agents generated over 1,500 merged pull requests, saving 60% to 90% on migrations, per Chief Architect Niklas Gustavsson.

Without platforms, AI amplifies friction. ‘Without a platform perspective, AI does not increase impact. It increases friction—and risk,’ warns devm.io. Platforms now treat AI models as deployable resources with role-based access, quotas, and governance, evolving IDPs to support ML teams and data scientists alongside developers.

Challenges persist: Agent interoperability lags, akin to early Open Banking, requiring standards like Model Context Protocol (MCP). Experts like Max Marcon of MongoDB predict cautious cross-organization agent collaboration by 2026.

DIY Fatigue and Maturity Pressures

Early enthusiasm for custom IDPs has waned. Roadie notes self-hosting Backstage—the dominant portal framework with 89% market share—takes 6-12 months for value, often stalling at low adoption around 10%. Maintenance diverts engineers from differentiators like golden paths.

‘Building a developer portal is not the same as building a platform. The portal is the interface; the platform is the substance behind it,’ Roadie emphasizes. Leaders now favor managed solutions to focus on unique integrations, measuring success via DORA metrics, SPACE frameworks, and developer surveys.

For VPs of engineering, priorities include software catalogs for service discovery, self-service templates, and data-driven scorecards to enforce standards, ensuring platforms evolve as products with roadmaps and user feedback.

Organizational Shifts and Governance Demands

Platform teams must navigate multi-disciplinary ownership—infrastructure, APIs, data, AI—encoding intent through defaults rather than mandates. PlatformEngineering.org stresses treating platforms like SaaS products, with dedicated product managers bridging personas from developers to ML engineers.

Security-by-design integrates FinOps and observability natively, as platforms mature per the Cloud Native Computing Foundation’s model. At easyJet, Platform Lead Molly Clarke advises: ‘Ask the users what they want.’ Spotify’s 55% drop in new developers’ time-to-first-PR post-Backstage exemplifies user-centric gains.

Governance looms large with agentic AI: Permissions, accountability, and drift detection become paramount, with GitOps extending as the control plane for infra and ops.

Predictions Shaping 2026 Deployments

Looking ahead, agentic platforms will embed AI across the software lifecycle, from planning to monitoring. PagerDuty’s João Freitas envisions agents making independent decisions with tools and resources. Yet fundamentals endure: No AI skips basics like testing or compliance.

The market reflects urgency, valued at $7.19 billion in 2024 and projected to hit $40.17 billion by 2032 per SNS Insider. Platform Engineering Week positions itself as the forum to tackle these threads, urging attendees to integrate before AI scales risks.

As Helen Greul of Multiverse.io cautions via The New Stack, amid tech acceleration, ‘people need steadiness.’ Successful organizations will balance innovation with structure, positioning platforms as the backbone for AI-driven value creation.