It’s time to bridge the technical gaps and cultural divides between DevOps, DevSecOps, and MLOps teams and provide a more unified approach to building trusted software. Call it EveryOps.
There are many moving parts in software development, particularly as tech and the role of engineers quickly evolve. Against this backdrop, there is transformative potential for “EveryOps” in 2025.
But what exactly is EveryOps? We coined the term to include DevOps, DevSecOps, MLOps, and any other additional Ops on the way. Here, we will examine how EveryOps will redefine the software development process while addressing current limitations and future needs.
The current state of software development
When discussing the operational aspects of this market, it becomes clear that software development is highly nonlinear. In the application development tools space, for example, there is a progression from basic applications to more advanced applications. The trend of applications becoming more complex is also impacted by the incorporation of AI into the newly developed software. This pattern is familiar from earlier shifts, like those seen with client-server models and cloud computing.
However, a third area of focus that includes elements like observability and CI/CD (continuous integration and continuous delivery) is particularly interesting. Many startups in this space have matured into later phases of their life cycle, and while some businesses maintain a clean, linear, supply-chain-like approach, larger enterprises in the domain may apply a different strategy. These companies often acquire various tools and technologies, leading to overlapping and sometimes conflicting components that they must integrate and reconcile.
This has raised questions about whether the operational space is collapsing. In reality, rather than consolidating neatly, the market remains highly fragmented. There is still a significant need to educate the industry on how new tools work, and new startups continue to emerge.
It’s a wild, fragmented space that presents both challenges and opportunities, and an EveryOps philosophy is a way organizations can make sense of it all.
EveryOps: Building trusted software
The concept behind EveryOps is centered around building trusted software. To achieve this, we advocate for a model resembling a traditional factory, complete with a supply chain—in this case, the software supply chain. At the end of the day, the goal for organizations is to operate a trusted software factory. They need to be able to inspect and prove the secure output of software components to various stakeholders such as compliance managers, CISOs, CIOs, and even external auditors. Sometimes, it’s simply about adhering to internal organizational policies to deliver software to customers, whether for a device or a service.
With this in mind, it doesn’t matter what components are included in the software you create or what runtime environment you ultimately deploy to. Some domains, like IoT, are still developing but will mature. For instance, we’ve had initial discussions about software in electric vehicles, and we anticipate these developments will extend even to remote or peripheral devices. This shift from servers to edge devices is inevitable. We also see this vision unfolding in areas like machine learning.
The core message of EveryOps is that a secure software factory must encompass everything. Automation is essential. Software pipelines must include robust policies to ensure trustworthiness. Ultimately, you must maintain control and demonstrate trust in the entire process.
For me, the secret sauce lies in building a system that is highly opinionated in its foundational concepts but remains open and API-driven. This approach allows organizations to gradually adopt and integrate different aspects of the software supply chain while ensuring security and reliability throughout.
Why EveryOps is imperative
There’s been a clear shift in software development towards developers owning applications end-to-end across the software development life cycle (SDLC), from coding and deployment to security and maintenance. The adoption of cloud platforms drove this trend, which aims to accelerate the delivery of secure, high-quality applications.
This DevSecOps approach shines by enhancing developer productivity and fostering collaboration within organizations that embrace DevOps principles. By baking security practices into every stage of the SDLC, DevSecOps reduces vulnerabilities and puts teams in a position to deliver high-quality software at an accelerated pace. Fast forward to today, and software development has evolved to incorporate increasingly complex dependencies, with new challenges emerging. This is especially true with the growing influence of machine learning, AI, and generative AI technologies.
AI and machine learning are no longer standalone initiatives limited to data science teams. They are becoming deeply embedded into modern software systems, making integrating MLOps into the broader DevOps and DevSecOps ecosystems necessary. Traditionally, MLOps practices operated in silos and focused on the needs of data scientists and engineers. As the boundaries between DevOps, DevSecOps, and MLOps blur, organizations need a more unified approach to effectively manage the complexities of all three domains.
This is why the EveryOps philosophy is vital in the modern age of software development.
Bridging gaps and building unity with EveryOps
Embracing EveryOps is a paradigm shift emphasizing the importance of bridging the technical gaps and cultural divides between DevOps, DevSecOps, and MLOps teams. It’s a holistic, inclusive approach to software and machine learning development, where ops teams are responsible for all domains across a unified EveryOps software supply chain and development pipeline. It’s all about creating a culture of end-to-end responsibility and continuous improvement so teams can maintain velocity without compromising trust or quality.
A core principle of EveryOps is seamlessly integrating trust and accountability into the SDLC with minimal friction for developers. This means security must be incorporated into the process and streamlined to avoid unnecessary bottlenecks. Developers, data scientists, and engineers must leverage tools and processes that empower them to work efficiently while adhering to best practices.
This sounds enticing, but what does it take in practice? At a high level, making EveryOps a reality means leveraging tools and frameworks designed for cross-collaboration between traditionally siloed teams. As we know, data scientists and engineers bring unique skills and requirements to the table, with functions including model training, data preprocessing, performance optimization, and more. Meanwhile, developers tend to prioritize aspects like CI/CD pipelines, infrastructure as code, and application scalability.
The convergence of these varied roles requires applying sound engineering principles to MLOps to ensure transparent management of a fully automated machine learning life cycle—from data preparation to model deployment and monitoring. Like traditional DevOps or DevSecOps workflows before it, centralized management of machine learning workflows and artifacts is critical for creating a unified view everyone can rely on.
Benefits of adopting EveryOps for machine learning
By embracing EveryOps, organizations can expect several key benefits, including:
- Enhanced trust: A consolidated visibility for machine learning workflows and artifacts allows stakeholders to quickly and confidently rely on the outputs of AI systems, knowing they’ve been developed and deployed with accountability.
- Improved efficiency: Streamlined, automated processes and shared tools minimize team friction, enabling faster cycles and more effective collaboration.
- Scalability and resilience: Unified EveryOps practices ensure that software and machine learning systems can scale effectively while maintaining top-level security and reliability.
- Cultural alignment: Breaking down silos creates a culture of end-to-end responsibility and continuous learning, which drives innovation and long-term success.
The EveryOps philosophy isn’t just a strategy for integrating DevOps, DevSecOps, and MLOps—it’s a call to action for organizations to embrace a unified, collaborative mindset that transcends technical and cultural barriers. As adoption of AI and machine learning continues to rise, EveryOps will be critical in ensuring organizations remain agile, secure, and competitive in an increasingly complex landscape.
The EveryOps philosophy is already redefining software development. By uniting DevOps, DevSecOps, MLOps, and emerging Ops under a cohesive framework, organizations give themselves a better chance at addressing the complexities of modern software and new machine learning workflows. Further, prioritizing trust, visibility, and automated controls across the software supply chain ensures teams can deliver secure, reliable, and scalable solutions.
Adopting EveryOps is not just an option—it’s imperative for staying competitive. Bridging gaps, fostering cultural alignment, and enabling speed and productivity empower organizations to thrive.
Yoav Landman is co-founder and CTO of JFrog.
—
New Tech Forum provides a venue for technology leaders—including vendors and other outside contributors—to explore and discuss emerging enterprise technology in unprecedented depth and breadth. The selection is subjective, based on our pick of the technologies we believe to be important and of greatest interest to InfoWorld readers. InfoWorld does not accept marketing collateral for publication and reserves the right to edit all contributed content. Send all inquiries to [email protected].
