devopsy.kinsta.cloud

Category: AI

How DevOps Powers Boundless Automation in Industrial Environments
In an era of digital transformation, automation is no longer a luxury — it’s a necessity. Industrial environments, once resistant to rapid technological change, are now embracing automation to streamline operations, reduce errors and enhance productivity. At the heart of this shift lies DevOps: A methodology that fosters collaboration, accelerates deployment and enables continuous improvement.

DevOps methodologies are driving boundless automation in traditionally complex operational technology (OT) settings. From manufacturing to utilities and logistics, DevOps professionals are uniquely positioned to bridge the gap between software and machines, transforming industrial workflows through smarter, more integrated systems.

The Role of DevOps in Industrial Automation

DevOps is built on principles of collaboration, automation, iterative improvement and shared responsibility. While these ideas originated in software development, their impact is now reshaping industrial environments as well.

In complex OT settings where machines, sensors and control systems operate 24/7, manual workflows and siloed teams can lead to inefficiencies and costly downtime. DevOps helps eliminate these roadblocks by applying the same automation-first mindset that revolutionized software delivery. Here’s how DevOps is transforming industrial automation:
- Continuous Integration and Deployment (CI/CD): Automated pipelines allow for rapid testing and deployment of industrial control software, firmware updates and configuration changes without disrupting production.
- Infrastructure as Code (IaC): OT networks, PLC configurations and device settings can now be defined and managed programmatically, ensuring consistency across facilities and eliminating human error.
- Monitoring and Observability: DevOps enables real-time monitoring of both digital systems and physical assets, allowing operations teams to spot trends, detect anomalies and act proactively.
- Feedback Loops: DevOps encourages short feedback cycles — essential for refining industrial automation strategies and minimizing operational waste.
Bulletproofing the Plant Floor With Automation

By integrating DevOps tools and culture into industrial settings, organizations unlock powerful automation benefits that drive productivity and reliability. Consider the following examples of key areas where automation excels:
- Machine Configuration & Provisioning
  Automate the setup and calibration of control systems, reducing deployment time and ensuring standardized environments across multiple sites.
- Data Collection and Analysis
  Collect data from industrial sensors and edge devices, such as handheld gas flow meters and feed it into centralized platforms for predictive maintenance, compliance reporting and optimization.
- Incident Response and Recovery
  Automatically detect and respond to system anomalies, minimizing downtime and preventing small issues from escalating into production halts.
- Safety and Compliance Audits
  Automate routine safety checks, environmental measurements and audit reporting for seamless regulatory compliance.
Case in Point: Empowering Maintenance Teams

Maintenance in industrial environments has historically been reactive. However, when DevOps is combined with smart asset management tools, it evolves into a proactive, data-driven discipline.

For example, teams can use IoT-connected devices like a handheld gas flow meter to gather real-time readings. If the flow levels are abnormal, the system automatically creates a ticket in the maintenance platform. It then cross-references the issue with historical data to determine the likely cause. Based on that analysis, a repair recommendation is generated and scheduled without human intervention. This kind of intelligent automation reduces downtime, improves safety, ensures compliance and extends the life of equipment.

Culture and Collaboration Are Key

Implementing DevOps in industrial settings isn’t just about tools — it’s about fostering a culture that supports transformation. Automation thrives when cross-functional teams work toward shared goals and openly exchange knowledge.

When DevOps professionals collaborate with OT teams, they help build a culture of shared accountability for uptime and performance. Together, they can plan automation initiatives in sprints, hold regular retrospectives to refine processes and continuously improve operations. These cultural shifts can be challenging, but they’re also where the most impactful change happens. Once IT and OT teams begin speaking the same language, innovation accelerates — and so does progress.

Overcoming Industrial Challenges With DevOps Thinking

Applying DevOps to industrial automation is not without its challenges:
- Legacy Systems: Many facilities still run outdated equipment that isn’t easily integrated with modern tools.
- Cybersecurity Concerns: Expanding networked automation increases the attack surface, requiring built-in security practices from the start.
- Skill Gaps: Traditional OT staff may be unfamiliar with DevOps concepts, requiring training and change management.
However, these challenges can be addressed through:
- Containerization and API gateways to wrap legacy systems
- DevSecOps practices that build security into every phase of deployment
- Cross-training programs to upskill operations staff on automation tools
The Future Is Automated and DevOps Is the Key

The industrial world is at a turning point. As demand for efficiency, flexibility and data-driven decision-making rises, automation becomes the path forward. And DevOps is the engine that powers that journey.

By applying DevOps principles — automation, collaboration, continuous delivery and real-time monitoring — practitioners can modernize industrial systems, eliminate silos and unlock powerful gains in uptime, quality and throughput.

For DevOps professionals, the opportunity is clear: Bring your expertise to the plant floor. Whether it’s integrating a handheld gas flow meter into a predictive maintenance workflow or building CI/CD pipelines for PLC software, your skills can help shape the next generation of industrial innovation.
May 30, 2025
Harness Adds MCP Server to Expose Data to Third-Party AI Tools

Harness today revealed it has developed a model context protocol (MCP) server through which it will be able to provide more context to artificial intelligence (AI) tools, agents and platforms that are being used to build software.

Rohan Gupta, a principal product manager for continuous delivery and AI DevOps for Harness, said exposing the data residing in the Harness continuous integration/continuous delivery (CI/CD) platform will streamline workflows by eliminating the need to rely on custom plug-ins and application programming interfaces (APIs) to provide additional context about the application development environments where code is ultimately deployed.

Originally developed by Anthropic, MCP is based on a JSON-RPC 2.0 API framework and is rapidly becoming a de facto API standard for integrating AI tools, agents and platforms. The overall goal is to make it easier to expose relevant data to any AI agent that has permission to access it. The overall goal is to make it simpler for AI coding tools to more easily understand the reasons why, for example, a specific DevOps pipeline might have failed, said Gupta.

DevOps teams will also be able to employ the role-based access controls (RBAC) provided by Harness to limit access to deployment logs only. All the API keys are managed through Harness and no data is ever sent to a large language model (LLM).

It’s not clear how widely AI tools and platforms have been adopted but a Futurum Group survey finds 41% of respondents expect AI tools and platforms will be used to generate, review and test code, while 39% plan to make use of AI models based on machine learning algorithms.

Ultimately, DevOps engineers will soon find themselves supervising a small army of AI agents with the help of what are becoming known as Planner agents, said Gupta. It remains to be seen how multiple AI agents will interact with one another to complete an end-to-end task but as orchestration frameworks continue to evolve there will be higher levels of interoperability between agents that both Harness and its third-party partners are continuing to build, said Gupta.

In the meantime, DevOps teams might want to start creating an inventory of manual tasks that might soon be assigned to an AI agent. As the ability of these AI agents to reliably complete these tasks improves, DevOps engineers should be able to spend more of their time on activities that provide higher value to the business, he noted.

Arguably, the next major challenge will be the cultural changes that will inevitably occur as more DevOps tasks are automated using AI agents that, for all intents and purposes, are now part of a DevOps team, added Gupta.

Regardless of approach to building software, the expectation is that the data that AI agents need to write better code will be readily made available. Each organization will need to decide to what degree to trust AI technologies to build software themselves, but ultimately, as the pace of software innovation starts to accelerate in the age of AI, it may soon not be feasible to effectively compete without it.

May 29, 2025
New Relic Integrates AI Agents with Copilot Coding Agent from GitHub

New Relic this week made available a preview of an integration with the artificial intelligence (AI) agents it created for its observability platform with the AI agents that GitHub has developed.

Nic Benders, chief technical strategist for New Relic, said AI agents that the company has developed will now be able to create an issue that GitHub Copilot will be able to use to improve an application. GitHub Copilot will analyze the issue, draft a fix, and submit a draft pull request for human review. The New Relic platform will then validate the change once it is merged.

In effect, the New Relic and GitHub agents are creating a virtuous cycle to address incidents and performance as they arise, said Benders.

Just as importantly, the New Relic agents are creating issues in a semi-structured format that is human readable, so DevOps teams will have the option of either assigning that task to a human developer or GitHub Copilot, he added.

New Relic is making this capability initially available to eligible accounts that are also Copilot Pro+ or Copilot Enterprise users. Launched at the Microsoft Build 2025 conference, the GitHub Copilot coding agent is currently available as a preview to GitHub Copilot Enterprise customers and GitHub Copilot Pro+ users.

While DevOps teams are making greater use of AI tools, it might be a few years yet before they are fully mastered, noted Benders. Using AI tools to generate code is one thing but integrating AI agents into DevOps workflows is a more complex endeavor, he noted.

Nevertheless, a recent Futurum Group survey finds 41% of respondents expect AI tools and platforms will be used to generate, review and test code, while 39% plan to make use of AI models based on machine learning algorithms. The challenge now is identifying the specific tasks that might best be handled by an AI agent to reduce the overall amount of toil that goes into software engineering, said Benders.

Ultimately, AI agents will also make DevOps teams less dependent on a confusing set of observability dashboards as AI agents begin to proactively discover issues, hopefully, before they have any significant adverse impact on application environments.

Each organization will need to determine for themselves the degree to which they are comfortable relying on AI to build and deploy applications, but at this point, it’s more a question of how much rather than if. There are clearly productivity gains to be had so long as the work of AI agents remains closely supervised by DevOps engineers that have the expertise required to ensure that application quality is maintained and enhanced.

In the meantime, DevOps teams should be aggressively experimenting with AI tools and platforms, said Benders. Like most powerful tools, it requires knowledge and expertise to get the most out of AI that only comes with hands-on experience, he added.

After all, the DevOps goal has always been to automate as many tasks and workflows as possible. AI agents are only the latest in a long series of tools and platforms that enable DevOps teams to achieve that goal.

May 29, 2025
AWS Application Modernization Service Employs AI Agents to Refactor Code

Amazon Web Services (AWS) has made generally available a service that makes use of artificial intelligence (AI) agents to migrate and modernize infrastructure, applications and code.

Sriram Devanathan, general manager for low-code/no-code AppBuilder service at AWS, said the AWS Transform service provides access to a natural language chat interface that makes it simpler to either refactor, for example, VMware, mainframe and .NET workloads or upgrade older instances of Java.

Initially previewed as an extension to Amazon Q Developer, AWS has a series of AI agents that have been specifically trained to both understand how code is constructed and invoke various tools, such as application assessment tools from CAST Services, to transform it. DevOps teams can use AWS Transform to specify objectives, share project context, assess business plans and cost savings, review and adapt transformation plans, identify dependencies, approve code infrastructure suggestions and execute tests.

For example, an AWS Transform agent for .NET can migrate a Windows-based application to Linux up to four times faster than traditional methods to reduce Microsoft cost by as much as 40%, noted Devanathan.

Similarly, an AWS Transform agent for VMware can automate the conversion of on-premises VMware network configurations to AWS equivalents up to 80 times faster than manual approaches.

An AWS Transform agent for mainframe can also decompose monolithic z/OS COBOL applications into components that can be run in the cloud in minutes. AWS Transform agents can, for example, invoke a Graph Neural Networks capability that makes it simpler to refactor applications written in COBOL, JCL (Job Control Language) and those relying on Customer Information Control System (CICS) transaction manager, BMS (Basic Mapping Support) screens, DB2 databases, and VSAM (Virtual Storage Access Method) data files in a set of Java-based microservices that can be deployed on an open source Postgres database.

In many cases, the documentation generated after that migration is more detailed than what was created when those legacy mainframe applications were initially built and deployed, said Devanathan.

Most organizations will still require specialists to migrate and modernize applications, but the total amount of time and effort required will be substantially reduced, said Devanathan. Most IT teams would be well-advised to revisit their modernization strategy by picking a relatively small project to better understand the level of refactoring that can now be achieved using AI agents, he added.

Longer term, AWS is looking to expand the range of modernization tools that its AI agents are trained to invoke, he added.

During uncertain economic times, the amount of pressure being applied to reduce the total cost of IT always rises. As such, interest in migrating legacy applications always tends to rise in an economic downturn. However, the amount of effort historically required to modernize applications has been substantial. In fact, many of these projects have never been completed simply because the cost of migrating applications proved to be much higher than initially anticipated.

Regardless of motivation, however, the total cost of continuing to run legacy applications can, in many cases, be even more prohibitive than the cost of a modernization initiative that, with the help of AI agents, may now prove to be a more viable option.

May 28, 2025
Claude Opus 4: The AI Revolution That Could Transform DevOps Workflows

The DevOps landscape just witnessed a seismic shift. Anthropic’s latest release, Claude Opus 4, isn’t just another incremental AI improvement — it’s positioning itself as the “best coding model in the world” with capabilities that could fundamentally reshape how DevOps teams approach their daily workflows. With the ability to work autonomously for up to seven hours, this AI assistant promises to tackle the complex, time-consuming tasks that have long plagued development and operations teams.

Beyond Code Generation: A True DevOps Partner

While many AI coding assistants excel at generating snippets or explaining syntax, Claude Opus 4 represents a paradigm shift toward comprehensive workflow automation. This extended operational capability opens unprecedented possibilities for DevOps professionals managing infrastructure, CI/CD pipeline optimization and system monitoring.

The seven-hour autonomous work window isn’t just about writing more code — it’s about sustained problem-solving across the entire DevOps lifecycle. Imagine deploying Claude Opus 4 to analyze your infrastructure logs, identify performance bottlenecks, draft remediation scripts and even document the troubleshooting process while you focus on strategic initiatives. This sustained AI engagement could transform reactive DevOps practices into proactive, intelligence-driven operations.

Infrastructure as Code Gets Smarter

One of the most compelling applications for Claude Opus 4 is infrastructure as code (IaC) development and maintenance. DevOps teams often struggle with complex Terraform configurations, Kubernetes manifests and cloud resource optimization. Claude Opus 4 could revolutionize how teams approach infrastructure automation with its enhanced coding capabilities and extended operational window.

The AI’s ability to think step-by-step through complex problems becomes particularly valuable when architecting scalable infrastructure solutions. Rather than generating YAML files or Terraform scripts, Claude Opus 4 can analyze your existing infrastructure, identify optimization opportunities and propose comprehensive solutions considering security, scalability and cost efficiency.

CI/CD Pipeline Evolution

Continuous integration and deployment pipelines represent another area where Claude Opus 4’s capabilities could drive significant improvements. Modern DevOps teams manage increasingly complex build processes, testing frameworks and deployment strategies across multiple environments. The AI’s sustained work capability means it could spend hours analyzing pipeline performance, identifying bottlenecks and implementing optimizations without requiring constant human oversight.

Consider the typical scenario where a deployment fails in production. Traditional troubleshooting involves manual log analysis, code review and iterative testing — processes that can consume entire work days. Claude Opus 4 could autonomously investigate the failure, analyze related system metrics, identify the root cause and even draft the fix, all while maintaining comprehensive documentation of the investigation process.

Monitoring and Observability Enhancement

The modern DevOps landscape generates massive monitoring data, alerts and performance metrics. Claude Opus 4’s analytical capabilities could transform how teams approach observability by automatically correlating data across multiple monitoring tools, identifying patterns that human operators might miss and generating actionable insights.

The AI could continuously analyze application performance metrics, infrastructure health data and user experience indicators to identify potential issues before they impact production systems proactively. This shift from reactive monitoring to predictive operations represents a fundamental advancement in DevOps maturity.

Security and Compliance Automation

Security remains a critical challenge in DevOps workflows, with teams struggling to maintain compliance while delivering features rapidly. Claude Opus 4’s enhanced coding abilities and sustained operational capacity could automate many security-related tasks, from vulnerability scanning and remediation to compliance reporting and security policy implementation.

The AI could analyze codebases for security vulnerabilities, review infrastructure configurations against security best practices and even generate security documentation — all while maintaining the detailed audit trails required for compliance frameworks.

The Claude Code Integration Advantage

The introduction of Claude Code, allowing direct terminal integration, represents a game-changer for DevOps workflows. This capability enables seamless integration with existing toolchains, allowing teams to leverage Claude Opus 4’s capabilities without disrupting established processes.

DevOps engineers could invoke Claude directly from their command line interfaces to troubleshoot issues, optimize configurations, or automate routine tasks. This level of integration brings AI assistance directly into the operational environment, where DevOps teams spend most of their time.

“Claude Opus 4’s capacity for prolonged, multistep reasoning and its ability to sustain complex workflows for hours is a significant step towards genuine agentic DevOps,” said Mitch Ashley, VP practice lead, software lifecycle engineering at The Futurum Group. “We’re moving beyond discrete AI task automation to AI systems capable of shouldering increasingly complex tasks, which will ultimately help us transform how we operationalize DevOps workflows, pipelines and CI/CD for DevOps and platform engineering.”

Preparing for AI-Enhanced DevOps

As Claude Opus 4 capabilities become available to DevOps teams, organizations should consider integrating AI assistance into their existing workflows. The key lies not in replacing human expertise but in augmenting team capabilities to handle increasingly complex infrastructure demands.

Teams should start by identifying repetitive, time-consuming tasks that could benefit from AI automation, while ensuring proper oversight and validation processes remain in place. The goal is to free human engineers to focus on strategic initiatives, architectural decisions and innovation while AI handles the heavy lifting of routine operations.

Claude Opus 4 represents more than an improved coding assistant — it’s a glimpse into the future of AI-enhanced DevOps operations. As teams integrate these capabilities into their workflows, we will likely see a fundamental shift in how development and operations teams approach their daily challenges, ultimately leading to more reliable, efficient and innovative software delivery processes.

May 23, 2025
Perforce Commits to Building Agentic AI Fabric for DevOps Workflows

Perforce Software this week committed to building an agentic artificial intelligence (AI) framework that will be embedded across its entire portfolio of DevOps tools and platforms.

Company CTO Anjali Arora said Perforce Intelligence will provide access to a set of AI agents that have been trained to automate a range of software engineering tasks that DevOps engineers will be able to orchestrate.

Additionally, Perforce is committing to using the same fabric to integrate its AI agents with other AI agents being developed by, for example, cloud service providers, she added.

Perforce, for example, has already added several AI agents to platforms such as Gliffy, a diagramming platform for Confluence, a Wiki-based collaboration platform for sharing documents provided by Atlassian.

Next month, AI agents will also be added to the Puppet Enterprise Advanced platform to provide access to Puppet Infra Assistant, a natural language chat interface that allows DevOps teams to quickly understand what is happening across the IT environments that have adopted the IT automation that Perforce acquired in 2022.

Via that interface, it becomes possible to know how something is working, what versions of modules or server operating systems are being used and whether something is in or out of compliance without needing to know the Puppet programming language.

Ultimately, the AI agents being developed by Perforce will make it simpler for DevOps teams to create and maintain workflows across a portfolio of DevOps tools and platforms that Perforce has acquired over the past few years, including BlazeMeter, an application testing platform, Delphix, a provider of data masking platform, and, most recently, Snowtrack, a provider of version control software used by application designers.

The overall goal is to eliminate the need for scripts, frameworks, or other legacy approaches to integration in favor of AI agents that share access to an orchestration framework. Those agents will be able to proactively adapt as changes are made across the entire software development lifecycle (SDLC), said Arora.

DevOps teams that have had early access to these AI agents have seen 50% efficiency gains while increasing test coverage by 20%, according to Perforce. In time, it’s apparent DevOps teams will be able to more effectively manage a much broader portfolio of applications, noted Arora. The challenge will be adapting existing DevOps cultures to include AI agents that software engineers are assigning tasks to complete, with the help of other AI agents and policy controls to provide governance functions and validations of output, she added.

It’s not clear to what degree DevOps teams have adopted AI, but research from the Futurum Group finds 41% of respondents expect generative AI tools and platforms will be used to generate, review and test code. A total of 39% also plan to make use of AI models based on machine learning algorithms.

Regardless of what type of AI technologies are adopted, hopefully, DevOps workflows and pipelines will soon be reconstructed in a way that is significantly less fragile as more scripts and connectors are eventually replaced by AI agents.

May 22, 2025
Rethinking AI Software Development With a Three-Tiered Framework

Artificial intelligence (AI) is transforming software development into a more efficient and augmented process, prompting developers and their peers to rethink how they produce value. Nearly 100% of the code will likely be supported, generated and/or tested by AI by the end of this year for new code, allowing developers to focus more on design, architecture and problem-solving as they offload routine coding tasks.

Despite growth in implementation, AI-driven software development is not a one-size-fits-all endeavor, and a tiered, robust framework is essential to maximize output while minimizing risk.

This can be done by following a three-tiered framework for AI-driven development, where each facet performs a specific function.

1. Rapid AI-Generated Solutions

The first tier prioritizes speed above all else, with the primary objective of delivering value quickly. AI generates code almost entirely autonomously to rapidly create prototypes, internal tools, or short-term solutions that solve immediate business problems. Quick-generated software is the developers’ playground, underpinned by high experimentation, low friction and an increase in opportunities for “vibe coding” (collaborating with AI agents to develop working applications in hours rather than weeks). Throw-away code becomes common. Long-tail support of application needs has now become a reality. There is less concern for scalability or long-term maintenance, as it can be recreated quickly.

2. Collaborative AI-Human Development

The second layer, focused on collaborative AI-human development, is ideal for larger business-critical software that demands more governance, reliability and collaboration between humans and machines. AI doesn’t fully take over the keyboard; it evolves into a trusted co-pilot that writes test cases, suggests code snippets, writes selected portions, flags vulnerabilities and helps with documentation. Some agentic automation is expected while human-in-the-loop is enforced. But crucially, human developers still drive architectural decisions, co-validate outputs and maintain accountability. Humans are expected to refine their context curation and intent framing skills to enhance alignment between expectations and outcomes. This tier reflects a partnership that blends AI’s speed with human judgment to strike a balance between innovation and risk.

3. High-Reliability, Long-Term Software

The top of the pyramid is reserved for the most strategic, long-lived, mission-critical systems — those underpinning enterprise-wide processes or customer-facing platforms. These solutions must be durable, secure and compliant, often for extended periods due to the sheer size of their codebases, or the human-life impact of their functionality. While AI has a place here, its role is on average more assistive than generative. Agentic ecosystems play a critical role in ensuring higher tiers of quality, security, scalability and maintainability. It might automate ancillary repetitive tasks or surface patterns in telemetry data, but the bulk of development remains human-centric. This software is about resilience and quality, not just results. Eventually, this category will also evolve, but we still have time. A lot more maturing of these new tools is needed.

Operational Shifts to Support AI Integration

Adopting a three-tiered approach is only one piece of the puzzle. Enterprises must also undergo meaningful operational change to scale AI-led development across these layers. Legacy delivery models rooted in static requirements and monolithic releases won’t suffice. Enterprises need to shift toward agile, product-centric agentic structures that emphasize rapid iteration, measurable outcomes and continuous feedback.

Equally important is the modernization of IT infrastructure. Agent-native architectures, modular platforms that can be easily swapped, non-dependency-inducing vendor solution usage and scalable pipelines are essential for supporting the dynamic needs of AI-powered tooling, as is a robust data strategy. AI’s effectiveness hinges on access to clean, contextualized and trustworthy data. This means that enterprises must invest in robust data governance, interoperability standards and mechanisms for real-time data sharing across teams. The more accessible and high-quality the data, the more valuable and accurate the results become. However, as I keep reminding my peers, dependency on high-quality data sets has been dramatically reduced with this new generation of models since the hard work has been encapsulated in prêt-à-porter models.

Lastly, the human element cannot be overlooked. Developers, testers and product teams require new skill sets, including context curation and intent framing, to work effectively in AI-augmented environments. From understanding prompt engineering and model behavior to interpreting AI-generated insights, continuous learning must be embedded in the culture to absorb the compounding effects of technological changes. AI will not necessarily replace developers, but developers who understand AI will outpace those who don’t.

Better, Faster and More Precise

What’s emerging is not a world where AI builds all the software, but one where software gets built faster, better and more precisely. Just as cloud computing redefined IT infrastructure over the past decade, AI is redefining the software lifecycle.

By adopting these AI-led software development strategies, enterprises can optimize today’s processes and future-proof their entire application ecosystem. They should foster a mindset of continuous learning and a lifelong commitment to innovation, a necessity for staying at the cutting edge of technology.

May 22, 2025
The Era of AI-Native Software: Why Retrofitting AI Won’t Work And How DevOps Must Keep Up

Integrating artificial intelligence (AI) onto legacy software is like trying to get an old flip phone to run the latest augmented reality (AR) apps. Yet, enterprises continue to take this approach, in the process encountering compatibility issues, sluggish performance and AI behavior which deviates from expectations.

We live in an AI-centric age, where organizations have AI integrated into at least one business function. Statista reports that global AI adoption soared to 72% in 2024 from 55% in 2023. Hence, it is no secret that AI is quickly becoming a mainstream business tool instead of remaining an emerging trend. However, Boston Consulting Group’s report reveals that 74% of companies are yet to gain a tangible return on investment from AI.

So, how do we reap the obvious benefits of AI, which the 4% of the companies that have built cutting-edge AI capabilities across functions are reaping? The answer is simple — AI-native software.

Understanding AI-Native Software

Building AI-native software entails leveraging AI and machine learning (ML) as core components in applications and systems from the start, rather than considering it as an afterthought in the architecture. Such software can learn and improve its performance over time.

What sets AI-native software apart is its ability to combine the trifecta of deep research, reasoning and Agentic AI. This combination enables systems to analyze vast datasets and reason through complex scenarios to determine optimal and personalized outcomes. It then collaborates dynamically with AI agents to enhance research and decision-making.

The key difference lies in the structure. Older systems are about retaining stability and depend on fixed, predefined workflows that produce expected results. AI-native apps refine and continuously improve to adapt based on learned information. This shift in design philosophy demands an entirely new DevOps approach, one that aligns with the changing nature of AI models.

Challenging Traditional DevOps

DevOps pipelines were automated to improve the integration, testing and deployment of code in the software development lifecycle. This set of workflows meets the needs of deterministic applications where updates follow a set and predictable plan. Nevertheless, AI-native applications need more than a simple code change. They require continual retraining of models, handling of complex data evolution, ensuring real-time performance and validating outcomes across dynamic environments. Without a holistic approach, these challenges can disrupt established pipelines and lead to inconsistent results.

The complexity of AI-native applications requires seamless cross-functional collaboration. Quality engineering ensures model validation and addressing issues like bias and accuracy, while delivery excellence focuses on smooth integration, testing and deployment. Data science optimizes model performance, and security teams safeguard models from threats while ensuring compliance. Business analysts translate AI insights into actionable strategies, and product managers ensure alignment with business goals. Together, these teams work to maintain model performance, scalability and reliability throughout the lifecycle.

At Persistent, we have witnessed how organizations struggle when retrofitting AI into legacy systems — dealing with outdated infrastructure that isn’t AI-ready, fragmented data silos and the difficulty of adapting existing DevOps pipelines to handle AI models. By rethinking software architecture with AI as a foundational element, we can help them navigate these hurdles seamlessly. This approach ensures that data is clean, integrated and accessible, enabling AI to perform its best. Incorporating feedback loops, data validation and continuous model evolution into the development lifecycle allows AI systems to remain reliable and adaptive as the application evolves. This shift has empowered clients to move beyond one-off implementations, enabling smarter, more agile decision-making while keeping systems flexible, scalable and future-proof.

These complexities of AI-native applications demand an advanced DevOps strategy — one that incorporates an agile approach to AI for operations (AIOps) and ML for operations (MLOps) to ensure that applications are built and released in an optimized, dependable state at all times.

Leveraging AIOps and MLOps

AIOps introduces real-time monitoring, anomaly detection and automated issue resolution. It ensures that AI-powered applications evolve correctly. For example, if an AI chatbot begins generating biased responses due to skewed training data, AIOps steps in to flag and correct the anomaly before it disrupts customer interactions.

Meanwhile, MLOps manages a model’s life cycle, including version control, automated model retraining and performance audits. Without MLOps, AI models risk stagnation, leading to degraded performance and ethical concerns such as data drift, bias or compliance violations.

Together, AIOps and MLOps form a strategic framework that oversees the entire lifecycle and integrity of AI models, ensuring that every decision and action are accurate, unbiased and continuously improving.

Rethinking CI/CD for AI

Traditional CI/CD pipelines need to be redesigned to focus on data and model performance rather than just the code. Unlike AIOps and MLOps, which maintain the ongoing health and evolution of the models themselves, the modern CI/CD evolution centers on the deployment pipeline, which ensures that as models are updated, the process remains agile and responsive to new data.

Automating retraining and robust validation within the deployment pipeline can enable businesses to keep their models as agile as their environments. The approach goes further by shifting the focus of continuous monitoring from simplistic uptime checks to real-time evaluation of accuracy, bias and other performance factors to ensure swift, well-informed action can be taken when needed.

At Persistent, we encourage this shift — moving from code-centric to data-centric pipelines. We guide organizations to integrate model training and validation into CI/CD, implement continuous monitoring and feedback loops to eliminate biases and embed explainability and interpretability into every layer of the stack.

This evolution in CI/CD is less about software deployment and more about intelligence deployment, where AI models are refined and optimized just as frequently as the underlying application code. This evolved approach safeguards the integrity of AI models and reinforces their ability to drive impactful, reliable decisions in a world that waits for no one.

Securing AI-Native Applications

Adopting AI-native software is not just a technological change; it is a redefinition of how companies tackle software engineering, security and organizational workflows. To have value-based AI-native systems, businesses need to adopt cloud-first, containerized environments for scalable AI workloads, migrate from traditional databases to deep learning and predictive data lakes and balance regulatory compliance with ethical AI development to ensure unbiased and transparent decision-making.

Security is another essential aspect to consider in AI-native applications. The presence of adversarial attacks, model manipulation and data poisoning necessitates an integrated effort from security, DevOps and data science teams. The responsibility of making sure AI models function well while ensuring fairness, security and legal compliance is swiftly becoming an issue that companies must deal with.

At Persistent, we are committed to developing AI-native software that not only meets the highest standards of quality and performance but also addresses the unique security, compliance and ethical concerns associated with these applications. Our delivery excellence and quality engineering center of excellence (COE) play a critical role in ensuring that our AI-native software is developed with security, compliance and ethics in mind. Toward this, they conduct threat modeling & risk assessments, outline & implement secure coding standards & practices, develop & implement compliance frameworks, collaborate with regulatory bodies and finally conduct regular audits and assessments.

Paving the Road to AI-Native Systems

Software development is rapidly evolving — from integrating AI to being fundamentally shaped by it. The next wave of innovation requires companies to invest in AI architects who can code and embed intelligence deep within software.

At the leadership level, fostering an agile and adaptive mindset is essential. We focus on strategic guidance — setting ethical and business boundaries while empowering AI systems to self-optimize. Success indicators will evolve from counting detected flaws or delivered functions to measuring autonomously enhanced outcomes, such as AI-driven engagement models reducing customer churn.

The future belongs to those who lead, not just adapt. AI-native development isn’t merely an advantage — it is the foundation of the next era of technology. Companies that move decisively today won’t just optimize efficiency; they’ll create new business models, revolutionize customer experiences and set the benchmark for intelligent software. As AI evolves, the true differentiators will be those who innovate boldly, embed AI as a core capability and shape a future where software doesn’t just enable businesses — it drives entirely new possibilities.

May 22, 2025
Why You Shouldn’t Forget Workflows With Agentic AI Systems

As Matthew McConaughey runs around an airport wishing he had used an AI agent to book his flight, we could be forgiven for assuming that the technology is now a well-understood and integral part of everyday life for most people.

Of course, that’s not true. In reality, it’s quite far from the truth, even for businesses, but that is not to say that things aren’t going that way. In the future, many of our decisions will be handled by agents, or at least heavily informed by them. But in areas where we want more control, we will likely retain decision-making power, following a programmatic approach or workflow.

When booking a vacation, for example, we could consider different countries, cities and situations, the cuisine, the weather and the culture. Or we could just get agents to put together some options based on our preferences. The third option would be a mix of the two, such as choosing where you want to go and allowing agents to choose and book activities — again, based on our preferences and hobbies.

The choice will come down to how much autonomy we want to retain. And it’s the same with AI systems. In this article, I’ll explain the move from one decision-making method to another and show why both are still valid and complementary.

How Models Think

Traditional models have limited knowledge. They are trained on a finite amount of data, which puts some restrictions on the kinds of problems they can solve. For example, the first public iteration of ChatGPT could only deliver answers to prompts based on the information available up to 2022, as its underlying LLM, GPT-3.5, was trained on data up to that point. This is also why you will often hear or read about models that have been fine-tuned for specific use cases, such as HR or legal, as these use cases require additional steps and data to tailor them to the use case.

These models can be used to underpin generative AI applications within an organisation. They can therefore broadly answer queries relating to the models that have been fine-tuned, such as general HR policies or legal practices, but cannot answer specific questions employees might have that relate to their circumstances or organisational data. To unlock this ability, we can create a workflow that enables models to access and retrieve data from proprietary systems and internal corpora of data. This is what is known as a Retrieval-augmented generation (RAG) system, but one that is programmatic, i.e., a human has built in the capabilities that enable the model to access systems outside of its training data to extend its capabilities.

With tool-augmented LLMs, it is therefore possible to create advanced workflows that are limited only by the modular elements developers choose to include, meaning the control logic is determined by a human.

An agentic system uses an LLM’s ability to reason to determine the steps required to complete a task or fulfill a request. The difference between this approach and programmatic workflows boils down to the amount of autonomy we want our AI systems to have.

Structured workflows involve a linear progression to deal with problems. In many instances, this is the most appropriate approach, as it guarantees that the system won’t go beyond what it has been programmed to do. However, as the reasoning capabilities of LLMs have continued to develop, it is now possible to hand the control logic of compound AI systems over to the LLMs themselves. This is thanks to “chain-of-thought” reasoning that enables an LLM to break down problems, e.g., prompts, into logical steps, taking a little longer to consider each stage so that a more reliable and accurate output is delivered.

Orchestrating Agentic Autonomy

Orchestration frameworks, such as LangGraph, are used to build in the stages of reasoning that agents use to complete their tasks. This can include a stage in which the agent considers its initial answer and determines whether to proceed with or change its output. Orchestration frameworks also build in the tools that agents can use to complete their tasks. A simple example might be providing a specific tool that retrieves live meteorological information. By giving the agent access to this tool, it can bring in this live information for the use case that is being served, such as booking activities on vacation. No one wants to be struggling through outdoor activities when there’s a storm raging, which is why an agent who can reason on the information they get from calling a weather tool is particularly useful.

These agents are built in line with a framework known as ReAct (‘Re’ for ‘reasoning’ and ‘Act’ for ‘acting’), which ensures the agent follows the Thought-Action-Observation reasoning paradigm. These stages mean the agent will first consider the tools it has at its disposal to return a satisfactory response, such as a weather app, then take action by retrieving the information, observing the situation from the raw data and then presenting the information in a user-friendly format following a further reasoning stage.

Agents are the new kid on the block, but it doesn’t mean they’re here to usurp traditional workflows. The two work rather well in conjunction. Workflows can hand off to an agent, or team of agents, to complete tasks, just as with the use case I outlined at the beginning of this article — and it’s no different in a business context. The reverse can also be true, with agents handing off to pre-defined workflows once a course of action has been suggested. This can then lead to further feedback loops to iterate and ensure accuracy and better decision making.

Ultimately, the amount of autonomy we build into a system will always be determined by the use case, which will, in turn, be determined by the exact amount of control that is required.

May 21, 2025
Red Hat Previews Series of Updates to Ansible Automation Platform

Red Hat today at AnsibleFest 2025 previewed support for the Model Context Protocol (MCP) developed by Anthropic within Ansible Lightspeed, a set of generative artificial intelligence capabilities based on large language models (LLMs) specifically trained by Red Hat and IBM to automate IT tasks.

Any LLM or associated AI agent will soon be able to invoke AAP to execute a playbook. Additionally, IT teams can make use of Ansible policy enforcement based on Open Policy Agent (OPA) framework to establish guardrails for any AI agent or LLM accessing MCP.

Red Hat also previewed a customizable dashboard for the Ansible Automation Platform (AAP) that makes it simpler to surface the economic value created by an automated workflow. IT teams will be able to ascribe a value to an automation that will then generate a dashboard showing its value based on how often it runs. That dashboard can then be saved as a PDF file or embedded directly into a spreadsheet application that can be shared with finance teams to help evaluate the return on investment in automation.

Finally, Red Hat and HashiCorp previewed additional integrations planned with the Terraform infrastructure-as-code platform following the acquisition of HashiCorp by IBM, the parent company of Red Hat.

Sathish Balakrishnan, VP and general manager of the Ansible business unit within Red Hat, told conference attendees there will soon be more out-of-the-box integrations to enable automated workflows that include Terraform IaC tools that are widely used to provision IT environments. The two platforms together provide a highly complementary set of capabilities for automating IT workflows on a more end-to-end basis, he added.

HashiCorp CTO Armon Dadgar, meanwhile, added there will soon be more Terraform hooks into AAP to provide deeper integrations into production environments.

It’s not clear how many DevOps teams are looking to integrate Terraform with Ansible, but the opportunity to automate IT processes more fully presents an opportunity to streamline workflows. The challenge, as always, is aligning what are often different teams within the same organization.

Regardless of the approach, it’s clear that as more workflows are automated, the overall level of toil experienced by IT professionals should sharply decline in the months ahead—especially as AI models are more broadly applied—even as the number of applications deployed in production environments continues to exponentially increase.

It’s not likely those AI models and agents will replace the need for DevOps engineers and IT administrators any time soon. However, the nature of those roles is likely to profoundly change as more tasks are assigned to AI agents that are supervised by an IT team.

In the meantime, those IT teams would be well-advised to start identifying which manual tasks being performed today might soon be automated using a combination of Terraform tools and AAP that share access to the same core AI model and agent technologies. Truth be told, chances are high the bulk of that list will be made up of tasks that no one really enjoyed doing in the first place.

May 20, 2025