2025 DevOps Predictions - Part 2
December 18, 2024

As part of DEVOPSdigest's annual list of DevOps predictions, industry experts — from analysts and consultants to the top vendors — offer thoughtful, insightful, and often controversial predictions on how DevOps and development technologies will evolve in 2025. Part 2 covers more predictions about software development.

CLOUD-NATIVE

Cloud-native platforms will become the standard for deploying new digital workloads. By 2025, most of the new digital workloads will be deployed on cloud-native platforms, offering greater scalability, flexibility, and efficiency.
Svetlin Nikolaev
Senior Director, DX UX & Emerging Products, Progress

WEBASSEMBLY

In 2025, DevOps will increasingly focus on cloud-native and serverless technologies, prioritizing performance, portability, security, low latency, scalability, and interoperability. WebAssembly is uniquely suited to these needs, offering near-native execution speeds and secure, cross-platform compatibility across cloud and edge. With support for multi-language interoperability and popular orchestration platforms, WebAssembly can deliver low-latency, scalable serverless functions for event-driven, distributed applications. Its evolving sandboxed architecture and interoperability will make it a foundational technology for the next era of serverless DevOps.
David Rant
Engineering Lead, Gearset


Decentralized Architectures

Decentralized protocols like Mastodon highlight a growing trend toward federated solutions that emphasize autonomy and adaptability. Enterprises can anticipate similar developments in AI training and inference, cryptocurrency, and event-driven architecture.
Robert Elwell
VP of Engineering, MacStadium

SELF-SUSTAINING DEV ENVIRONMENT

By 2025, engineers will shift their focus from low-level implementation tasks such as writing HCL configurations and YAML pipelines to higher-value activities in the areas of design, architecture, and performance-driven deployments. As infrastructure becomes increasingly as ephemeral as Docker containers, the entire engineering stack from infrastructure to code deployment will evolve from a "delicate" ecosystem requiring constant management to a more self-sustaining environment. This transformation will enable engineers to focus on the Software Development Life Cycle (SDLC) of value-driven workloads versus getting involved in cluster management, infrastructure deployment and day-zero operations. The result will be increased productivity and innovation in line with market demands and technological advancements.
Ben Ghazi
Co-Founder, Codiac

INFRASTRUCTURE AS CODE

In 2024, only 13% of organizations reported achieving IaC Maturity in Stacked Up: The IaC Maturity Report. IaC is an important tool that organizations need to ensure cloud native environments can scale and remain agile. In 2025, we will see more companies investigate infrastructure automation and management tools that can help them improve on their existing IaC investments. Infrastructure from Code will become a viable option for platform engineers as they seek to improve the developer experience and velocity organizations require.

In 2024, 97% of respondents in Stacked Up: The IaC Maturity Report reported difficulties with Infrastructure as Code (IaC) technologies, in part due to challenges with ensuring consistency both during IaC creation and maintenance. In 2025, we'll see organizations invest in infrastructure management and automation to overcome these IaC challenges. New tools that automatically enforce best practices for consistency, security, and scalability will make IaC faster and more streamlined, and reduce developers' cognitive load.
Sachin Aggarwal
Co-Founder and CEO, Stackgen

INFRASTRUCTURE FROM CODE

In 2024, nearly half of developers surveyed in Stacked Up: The IaC Maturity Report viewed IaC as a necessary evil required to deploy the applications they were hired to design and deliver. To improve the developer experience, many organizations will turn to Infrastructure from Code (IfC) in 2025 to allow developers to focus on the application, not the infrastructure.
Sachin Aggarwal
Co-Founder and CEO, Stackgen

SIMPLIFIED STATE MANAGEMENT

Developers will embrace simplified state management: In 2025, developers will seek alternatives to event-driven architectures as the complexities of fragmented workflows and debugging challenges continue to grow. Teams will adopt frameworks that centralize business logic and automate state management, which will reduce operational overhead while improving reliability and scalability. This shift towards durable execution will streamline the development of distributed systems and empower teams to focus on delivering value while ensuring robust and maintainable architectures.
Drew Gorton
Senior Director of Developer Relations, Temporal

KUBERNETES

In 2025 Kubernetes will remain the leading platform for compute workloads, with significant growth driven by GenAI workloads. As GenAI usage continues to grow, both in terms of scale and capabilities, companies will increasingly rely on Kubernetes for scalable infrastructure, solidifying its role as the go-to platform for running compute and GPU workloads, regardless of the industry. Simultaneously, enterprises will continue migrating workloads from on-prem to the cloud, shifting away from legacy systems and embracing modern architectures to reduce maintenance costs. This shift, combined with GenAI adoption for enhanced automation will enable improved system utilization, cost reductions, faster deployments, and integration of new technologies, making Kubernetes even more effective for ML development, scaling, and delivery of applications.
Aviram Levy
Tech Evangelist, Zesty

CENTRALIZED KUBERNETES MANAGEMENT

As more mission critical applications are moving to Kubernetes, organizations will move towards centrally managing Kubernetes, rather than allowing developers to manage their own, to increase security and lower costs.
Tobi Knaup
VP, GM of Cloud Native, Nutanix

CLOUD AGNOSTICISM

Cloud-Agnoticism Gains Momentum: Organizations are adopting hybrid and multi-cloud strategies to reduce vendor dependence, driving demand for "run anywhere" solutions like Kubernetes.
Robert Elwell
VP of Engineering, MacStadium

VOICE INTERFACES

Voice is on the rise. Developers, often early adopters, are increasingly writing code through voice interfaces. Thanks to advancements in AI, voice recognition has improved dramatically, positioning it for mainstream adoption by 2025.
Ryan Janssen
CEO, Zenlytic

OUTSOURCING DEVELOPMENT

Reduced barriers to onboarding will lead enterprises to outsource more software development roles: In 2025 we'll see enterprises shift back to outsourcing more software development roles to reduce the strain on internal teams and alleviate skills shortages, spurred on by technologies that reduce barriers to onboarding. For example, enterprises may decide to enlist an IT services provider with specialist data science skills to support the development of new AI use cases. This will require less time and fewer resources than re-skilling existing developer talent, or hiring full time data scientists whose skills may not be needed in the long term.

Outsourcing development roles has previously been more difficult due to lengthy onboarding processes. Typically, outsourced staff would spend months gaining access to tools and data, understanding workflows, and getting to grips with an organization's existing code base. In 2025, advances in GenAI, Cloud Development Environments (CDEs), and Internal Developer Portals (IDPs) will enable both new hires and outsourced staff to get up to speed much faster, so they can deliver value from day one.
Nick Durkin
Field CTO, Harness

Go to: 2025 DevOps Predictions - Part 3

Share this

Industry News

October 22, 2025

Platform Engineering Labs announced the launch of formae, an open source infrastructure-as-code (IaC) platform designed to address fundamental insufficiencies of tools like Terraform.

October 22, 2025

Snyk announced the launch of Evo by Snyk, an agentic security orchestration system designed to secure AI-native applications and tools, including GenAI and agentic.

October 22, 2025

Kong announced Konnect Metering & Billing, powered by OpenMeter, a new solution designed to enable organizations to unlock new revenue streams by productizing and effectively billing for their APIs, AI models, and data streams, turning digital assets into significant sources of income.

October 21, 2025

GitLab announced its 18.5 release, delivering new specialized agents, security insights capabilities, and a modernized user experience with quick access to GitLab Duo.

October 21, 2025

Couchbase announced Couchbase 8.0, delivering end-to-end AI data lifecycle support for enterprises building AI applications and agentic systems.

October 21, 2025

Kong announced the general availability of the latest update to Konnect Service Catalog, which now includes native integration with AWS API Gateway.

October 21, 2025

Sonar announced the launch of SonarSweep™, a service designed to improve the training of LLMs optimized for coding applications.

October 20, 2025

Kong announced that Kong Event Gateway will be available Q4 2025 as part of Kong Konnect, the unified API and AI platform.

October 20, 2025

Waydev announced the launch of Waydev AI, an AI-native conversational platform designed to transform how engineering leaders measure performance and understand AI's impact on delivery.

October 16, 2025

Coder introduced Blink in Early Access.

October 16, 2025

Kong announced the native availability of Kong Identity within Kong Konnect, the unified API and AI platform.

October 15, 2025

Amazon Web Services (AWS) is introducing a new generative AI developer certification, expanding its portfolio for professionals seeking to develop their cloud engineering skills.

October 15, 2025

Kong unveiled KAi, a new agentic AI co-pilot for Kong Konnect, the unified API and AI platform.

October 15, 2025

Azul and Cast AI announced a strategic partnership to help organizations dramatically improve Java runtime performance, reduce the footprint (compute, memory) of cloud compute resources and ultimately cut cloud spend.