devopsy.kinsta.cloud

Category: AI

VS Code’s Open Source AI Revolution: A New Chapter for Developers

The development landscape has just undergone a dramatic shift. Microsoft’s VS Code team has taken a bold step toward democratizing AI-powered coding by open-sourcing its GitHub Copilot Chat extension under the MIT license. This milestone represents more than just another code release; it’s a fundamental reimagining of how AI tools should be developed and deployed in the developer ecosystem.

Breaking Down Barriers to Innovation

The decision to make VS Code an open-source AI editor addresses a growing concern in the developer community: The black-box nature of AI coding assistants. Until now, developers have been using powerful AI tools without understanding their inner workings, from prompt engineering techniques to context handling strategies. This transparency gap has hindered innovation, making it challenging for developers to fully optimize their AI-assisted workflows.

By open-sourcing the Copilot Chat extension, Microsoft has lifted the veil on these critical components. Developers can now examine exactly how agent mode functions, understand what context gets sent to large language models and study the prompt engineering strategies that make these tools effective. This level of transparency transforms AI coding from a mysterious process into something developers can understand, modify, and improve.

Community-Driven AI Development

The open source approach leverages the same community dynamics that made VS Code itself a success. Over the past decade, VS Code has evolved from a Microsoft project into a community-driven platform, with thousands of extensions and contributions from developers worldwide. Now, the VS Code team is applying this proven model to AI functionality.

This community-centric approach promises to accelerate innovation in ways closed development cannot match. When developers can see precisely how AI features work, they can identify areas for improvement, fix bugs, and adapt functionality to meet specific needs. The collective intelligence of the open-source community often surpasses what any single organization can achieve, regardless of its resources.

Practical Benefits for Development Teams

The immediate practical implications are significant. Development teams can now audit AI functionality to ensure it meets their security and compliance requirements. Organizations with strict data governance policies can examine exactly what information is transmitted to language models and modify their behavior accordingly.

For individual developers, this transparency enables more effective prompt engineering and a better understanding of AI behavior. Instead of treating AI coding assistants as unpredictable tools, developers can study successful patterns and apply them more systematically. The ability to examine system prompts and implementation details provides insights that can improve AI-assisted productivity across entire teams.

“This marks Microsoft’s follow-through to open source GitHub Copilot Chat, announced at Microsoft Build 2025,” said Mitch Ashley, VP practice lead of software lifecycle engineering at The Futurum Group. “GitHub Copilot Chat serves as the direct interface between the developer and the underlying AI code generation and agent capabilities available through VS Code and GitHub Copilot Agent. This transparency helps build trust with developers as AI takes on increased workloads for developers.”

Technical Architecture and Integration Strategy

The VS Code team’s technical approach demonstrates thoughtful planning for long-term sustainability. Rather than keeping AI functionality in a separate extension indefinitely, they plan to integrate components directly into VS Code’s core. This integration strategy ensures that AI becomes a first-class citizen in the editor, rather than an afterthought.

The decision to maintain issue tracking in the central VS Code repository further reinforces this integration philosophy. It signals that AI functionality isn’t a side project but a central part of VS Code’s future direction. This architectural approach provides a foundation for more sophisticated AI features while maintaining the editor’s performance and reliability standards.

Industry Implications and Future Outlook

This move represents a broader shift in how technology companies approach AI development. As AI becomes integral to software development workflows, the traditional closed-source model creates dependencies that many organizations find uncomfortable. Microsoft’s decision to embrace transparency could pressure other AI tool providers to follow suit.

The implications extend beyond individual developer productivity. Educational institutions can now use real AI implementations to teach prompt engineering and AI integration techniques. Researchers gain access to production-quality AI coding systems for academic study. Security researchers can audit AI tools for potential vulnerabilities or biases.

Challenges and Considerations

Open-sourcing AI functionality presents its challenges. The VS Code team must strike a balance between transparency, performance optimization and consistency in user experience. Managing community contributions while maintaining code quality requires significant investment in review processes and documentation.

The partial approach, open-sourcing chat functionality while keeping inline completions closed, reflects these complexities. This staged rollout allows the team to learn from community feedback before expanding the open-source scope to more complex features.

The Path Forward

The VS Code team’s roadmap demonstrates a commitment to gradually expanding open-source AI functionality. Plans to provide inline completion capabilities through the open-source extension demonstrate long-term dedication to the transparency vision. A partnership with the broader open-source AI community suggests that the success of this initiative depends on ecosystem-wide collaboration.

This milestone represents more than a technical achievement; it’s a fundamental shift toward more democratic, transparent and community-driven AI development. As artificial intelligence reshapes software development, having open, auditable and modifiable AI tools becomes increasingly important for maintaining developer agency and innovation.

The future of coding is undoubtedly AI-assisted, but Microsoft’s approach ensures that the future remains open, transparent and community-controlled. For developers who value understanding their tools and shaping their evolution, this represents a significant step toward a more empowering development environment.

July 2, 2025
Kusari Adds AI Security Tool to Inspect Code as Pull Requests Are Made

Kusari has added an artificial intelligence (AI) tool that runs a security risk assessment every time an application developer makes a pull request.

Company CTO Mike Lieberman said Kusari Inspector is designed to make it simpler for security to be shifted left toward application developers in a way that doesn’t impinge productivity.

Most application developers care about security, but getting the information they need to ensure it has historically been problematic, said Lieberman. Kusari Inspector makes use of AI and dependency graph analytics to surface issues long before vulnerabilities and other structural flaws are incorporated into a software build, he said.

Using a tool priced at $10 per seat per month, Kusari Inspector continuously learns about the codebase and individual application developer preferences, making it possible to generate instant, context-rich, annotated security reports with inline explanations for every new or updated pull request. Developers are then provided go/no-go guidance, remediation suggestions, and step-by-step instructions to mitigate risks such as exposed credentials, misconfigurations, typosquatted or maliciously named dependencies and prohibited licenses.

In addition to identifying those issues, Kusari Inspector will also rank risky, low-trust, or vulnerable dependencies—direct and transitive—based on multiple data sources to provide developers with a deeper understanding of application security issues, said Lieberman. Developers via a natural language interface can also chat with Kusari Inspector to clarify findings, customize recommendations, and set security standards.

Finally, Kusari Inspector will automatically generate and collect source data for a software bill of materials (SBOM) across all connected projects and repositories.

The core Kusari platform leverages AI to identify dependencies and understand the blast radius of incidents using metadata it collects from throughout the software supply chain. That capability is now being extended further left to provide application developers with a tool that enables them to write more secure code.

It’s not clear how much DevOps teams are investing in AI, but a recent Futurum Group survey finds that 41% of DevOps professionals expect generative AI tools and platforms will be used to generate, review and test code. A separate Futurum survey of cybersecurity leaders finds all are investing in software supply chain security, with application security posture management (ASPM) and DevSecOps automation and orchestration topping the priority list, followed closely by security composition analysis (SCA) tools, application programming interface (API) security and dynamic application security testing (DAST) tools. In addition, 30% of respondents expect to be piloting a software bill of materials (SBOM) initiative in the next 24 months/

However, the source of the funding for these initiatives is becoming more of a shared responsibility, with only 21% of respondents reporting that security budgets are the sole source. In fact, half of the respondents (50%) noted that application development teams now own responsibility for application security.

Overall, only 25% of respondents said there is limited collaboration with application development teams, resulting in occasional friction, compared to 59% that said there is good collaboration with room for improvement. Only 16% said there is a tight partnership based on shared goals.

Hopefully, AI tools will continue to close that gap in a way that reduces the current level of stress felt today by all concerned.

June 30, 2025
Next Version of Grok Includes Advanced Coding Assistance: Reports

The newest Grok is expected to feature advanced coding assistance, a sign that owner Elon Musk is re-evaluating the chatbot’s strategy after Microsoft Corp. and OpenAI released powerful new models that set higher benchmarks for language model performance.

Musk’s artificial intelligence (AI) company xAI is skipping Grok 3.5 and moving straight to Grok 4. The latest iteration, expected to launch sometime after July 4, will likely focus on enhanced coding assistance and seamless integration with native code editors, making it a more powerful tool for developers.

Clues in the codebase of the Generative AI chatbot suggest xAI is building a native code editor within the Grok web interface, modeled after Visual Studio Code (VSCode), a source code editor developed by Microsoft. This would let Grok directly modify, write, or debug code, signaling a move towards “agentic coding” in which AI suggests code and acts more autonomously within an integrated development environment.

Exact implementation of the new approach remains uncertain, but would position Grok more competitively alongside AI assistants that provide deep integration for developers.

“Grinding on @Grok all night with the @xAI team. Good progress. Will be called Grok 4. Release just after July 4th. Needs one more big run for a specialized coding model,” Musk said on social-media platform X on June 27.

Grok 4’s updates indicate a move toward more sophisticated, integrated development tools driven by AI. For xAI, the shift fits into a larger plan to remain at the forefront of GenAI, as well as a key cog in xAI’s ambitious expansion to build the world’s most powerful AI cluster. The company has a roadmap to deploy 1 million GPUs for computational power.

Ultimately, Grok 4 aspires to dominate in complex reasoning tasks, with targeted performance improvements designed to outperform rivals like ChatGPT-4 Turbo in code generation, debugging, and technical problem-solving, according to various reports.

While details remain somewhat sketchy on the forthcoming Grok 4, Musk has made no secret of where he intends to take content and data on the chatbot.

The makeover will also reportedly be politically influenced: Musk has tasked Grok’s development team with excising “politically incorrect, but nonetheless factually true” info from its data banks to stop the app from providing answers that Musk disagrees with. Among its various claims that have raised his ire is its conclusion that Musk is “the biggest spreader of misinformation on X.”

June 30, 2025
Revolutionizing CI/CD: A Framework for Integrating Generative AI Across the Software Delivery Lifecycle
As a DevOps expert, I’ve witnessed firsthand the transformative power of automation and continuous improvement. Today, I’m excited to share a framework that takes our practices to the next level by integrating generative AI into the software delivery lifecycle. This approach not only streamlines our processes but also enhances the quality and efficiency of our deliverables.

The Gen AI-Powered CI/CD Framework

Here’s a step-by-step guide to building and implementing a generative AI framework for CI/CD:

Automated Code Generation and Review
1. Implement GitHub Copilot or Amazon CodeWhisperer for AI-assisted coding.
2. Use AI-powered static code analysis tools like SonarQube or Amazon CodeGuru Reviewer.
Action: Integrate these tools into your IDE and Git workflows, such as GitHub Actions.

Intelligent Test Generation
1. Leverage tools like Diffblue Cover for automated test case generation.
2. Use AWS DevOps Guru to identify areas needing additional testing based on historical data.
Action: Implement AI-generated tests in your test suites and CI pipelines.

Dynamic Infrastructure Provisioning
1. Utilize AWS CloudFormation or AWS CDK with AI-enhanced templates.
2. Implement predictive scaling using Amazon EC2 Auto Scaling with ML-powered forecasting.
Action: Create AI-driven IaC templates and integrate them into your deployment processes.

Automated Security Scanning
1. Integrate Amazon Inspector for AI-powered vulnerability assessments.
2. Use AI-enhanced SAST/DAST tools like Snyk, Checkmarx, or Veracode.
Action: Implement these scans in your CI/CD pipeline, triggered by code commits.

Intelligent Deployment Strategies
1. Implement AI-driven canary deployments using AWS CodeDeploy.
2. Utilize Amazon SageMaker to build ML models for predicting deployment success.
Action: Integrate these predictive models into your deployment decision-making process.

AI-Enhanced Monitoring and Alerting
1. Leverage AWS CloudWatch with AI-powered anomaly detection.
2. Implement predictive alerting using tools like Datadog or New Relic’s AI capabilities.
Action: Set up AI-driven alert thresholds and automated incident response workflows.

Continuous Feedback and Optimization
1. Use AI to analyze user feedback and application logs (AWS Comprehend for sentiment analysis).
2. Implement ML models to correlate code changes with performance metrics.
Action: Create dashboards that visualize AI-derived insights for continuous improvement.

Documentation and Knowledge Management
1. Utilize GitHub CoPilot or AWS CodeWhisperer to generate and update documentation automatically.
2. Implement AI-powered chatbots for internal knowledge sharing.
Action: Set up automated documentation pipelines triggered by code or configuration changes.

Implementation Strategy
1. Assessment: Evaluate your current CI/CD pipeline and identify areas where AI can add the most value.
2. Tool Selection: Choose AI-powered tools that integrate well with your existing stack. Consider AWS services for seamless integration if you’re already in the AWS ecosystem.
3. Pilot Project: Start with a small, non-critical project to test your AI-enhanced CI/CD framework.
4. Training: Upskill your team on AI concepts and the new tools you’re introducing.
5. Gradual Roll-out: Implement the framework in phases, starting with code generation and review, then moving through testing, deployment and monitoring.
6. Measure and Iterate: Use metrics like deployment frequency, lead time and change failure rate to assess the impact of your AI-enhanced pipeline.
7. Scale: Once proven effective, scale the framework across all your projects and teams.
Challenges and Considerations
- Data Privacy: Ensure that your AI tools comply with data protection regulations.
- AI Bias: Regularly audit your AI models for biases in code suggestions or deployment decisions.
- Over-reliance: Maintain a balance between AI assistance and human expertise.
Conclusion

Integrating generative AI into your CI/CD practices isn’t just about adopting new tools — it’s about reimagining the entire software delivery lifecycle. This framework provides a roadmap for leveraging AI to enhance every stage of your pipeline, from code generation to deployment and monitoring.

By implementing this framework, you’re not just optimizing your current processes; you’re future-proofing your DevOps practices. The combination of AI-driven insights and human expertise will lead to faster, more reliable software delivery and ultimately, better products for your end-users.
June 30, 2025
Why Software Migrations Fail: It’s Not the Code
Engineers often overlook the human and organizational dimensions of software migrations. They typically perceive these projects as technical undertakings, but institutional knowledge, communication gaps, team morale and resistance to change can significantly influence timelines and outcomes. Examining common pitfalls, effective strategies and emerging trends reveals a comprehensive view of the migration process from a human-centered perspective.

Recent research continues to validate that delays in software migration are the rule rather than the exception. In a 2023 McKinsey & Company report, over 70% of large-scale digital transformations fell short of timeline or return on investment (ROI) expectations, with a lack of cross-functional alignment and insufficient change management cited as leading causes. Similarly, a 2024 Boston Consulting Group survey revealed that nearly half of global C-suite executives reported that over 30% of their organization’s technology development projects suffered delays or budget overruns.

Understanding Software Migrations as a Human-Centered Process

Software migration involves transitioning systems, applications, or infrastructure from one environment to another. These transitions may include moving from on-premises deployments to cloud platforms, retiring legacy systems in favor of modern stacks, or switching technology between vendors. Despite their technical appearance, such migrations cause ripple effects throughout organizational workflows and operations.

These migrations often require retraining end users, maintaining legacy systems during the transition and navigating competing priorities. The disruption can stall development cycles and introduce operational risk. Overlooked in most planning stages is how much these projects rely on people’s undocumented knowledge, assumptions and behavioral responses.

Human and Organizational Barriers That Derail Migration Efforts

Scope creep or technical hurdles are frequently blamed as the cause of migration delays, yet deeper analysis reveals that resistance to change and knowledge gaps within teams are often more significant obstacles. End users familiar with a particular interface or workflow may be reluctant to adopt a new one. Internally, this resistance is not always voiced directly but manifests as disengagement or low adoption rates. Business stakeholders, too, may not fully grasp the strategic rationale behind a migration effort. Executive alignment weakens when the justification isn’t well articulated and cross-departmental cooperation becomes harder to sustain.

Undocumented institutional knowledge is another persistent challenge. Many decisions about data structures, hard-coded values, or interface dependencies exist only in the minds of long-tenured employees. When these employees move on or get reassigned mid-project, teams can lose the rationale behind legacy implementations. This knowledge vacuum causes surprises during implementation, especially when hidden dependencies emerge.

Managing Mid-Migration Surprises and Mitigating Risk

No matter how well-scoped a project is, assumptions made by different teams often go unspoken. Teams often operate in silos, assuming shared understanding when none exists. This misalignment can result in components built on conflicting premises, only discovered during testing or go-live. Mid-migration surprises include unanticipated manual workarounds, deprecated features that remain mission-critical and inconsistent definitions of business logic across systems. Manual processes known only to a few employees become barriers when they cannot be replicated in the new environment and uncovering these issues late in the process leads to delays and rework.

Incorporating discovery sprints before execution has proven effective in managing these risks. These sprints create space to unearth undocumented logic, validate inter-team assumptions and highlight edge cases early. Including a buffer—typically 10–20% of the timeline—also reduces the risk of missed deadlines.

Strategies to Reduce Rework and Support Successful Transitions

While tooling plays a supporting role, successful migrations hinge on clear communication and iterative execution. Automated migration tools can minimize manual tasks and streamline repetitive processes, such as data copying or schema translation. Relying solely on automation overlooks the need for continuous stakeholder alignment, however.

Here are some ways to reduce rework:
- Cross-team workshops. Use workshops to identify assumptions early and establish shared understanding.
- Regular check-ins. Catch evolving issues before they escalate by employing check-ins throughout migration.
- Phased rollouts. Transitioning rollouts over time rather than all-at-once helps organizations validate changes incrementally and apply insights from one phase to the next.
Communication breakdowns lead to duplicate efforts, reactive firefighting and coordination overhead. Maintaining old and new systems requires continuous attention and introduces operational drag. Factoring in learning curves and user adaptation time into project schedules can help set more realistic expectations. Offering flexibility during migration is equally vital for maintaining morale and sustaining trust.

Future Trends and Measuring Human Impact

Recent innovations are changing the migration approach. AI-powered tools now enable faster and more accurate transformations. In one notable case, Airbnb completed a migration, estimated initially to take two years, in just six weeks using AI tooling. These platforms automate repetitive code translations and configuration tasks, reducing reliance on large teams while preserving accuracy.

Low-code and no-code migration platforms simplify transitions by offering visual interfaces and guided workflows. These innovations allow a broader set of employees to participate in the migration process, increasing organizational resilience.

While AI accelerates execution, it does not eliminate the need for human oversight. Engineers now spend less time on rote tasks and more on designing, monitoring and validating migration paths. Prompt engineering, anomaly detection and performance monitoring are emerging as core responsibilities in this new paradigm.

Tools like GitHub Copilot and Amazon CodeWhisperer are increasingly integrated into migration workflows to assist with repetitive coding tasks, generate test cases and validate syntax conversions. These tools reduce the cognitive load on developers and help teams move faster through lower-level tasks, freeing up human resources to focus on architecture decisions, validation and user-centric improvements.

Metrics That Reveal Human Impact

Quantifying the human side of migration is complex but not impossible. Several indicators can provide insight into where people are struggling and may need additional support:
- On-call ticket volume. A spike in support requests during the migration window often reflects friction points, such as unclear interfaces, poor documentation, or insufficient training.
- Employee satisfaction scores. More extensive morale trends can reveal team emotional fatigue or frustration. Measuring satisfaction at different stages of migration provides valuable feedback loops.
- Attrition rates. Higher turnover among technical staff or support teams may signal disengagement, burnout, or disillusionment with the process.
- User feedback on new systems. Capturing qualitative input from end users via surveys, interviews, or feature usage can help determine whether the migrated tools effectively serve their needs.
- Project timeline adherence. Deviations from planned schedules often point to organizational misalignment, unanticipated rework, or breakdowns in communication.
These metrics collectively paint a more nuanced picture than completion percentages or delivery milestones. They help teams identify where to adjust code, expectations, workflows and morale.

Redefining Successful Software Migration

Technical upgrades may motivate software migrations, but the real complexity lies in navigating organizational behavior, knowledge gaps and human responses to change. Recognizing this dimension transforms software migration planning and execution. It’s not enough to have the right tools—teams benefit by fostering alignment, surfacing assumptions early and communicating continuously.

The future of software migration lies not just in AI and automation, but in embracing the human elements that make these transitions sustainable. As platforms evolve, the ability to manage systems and people will be the defining skill of successful engineering leaders.

Disclaimer: The opinions expressed in this article are those of the author. They do not purport to reflect the views of his employer.
June 30, 2025
Harnessing AI and Automation for the Future of Innovation in DevOps

In today’s fast-paced technological landscape, the transformative power of AI and automation is more evident than ever. As startups drive economic development with innovative ideas and relentless hard work, understanding and leveraging these technologies can significantly enhance their journey. This article delves into the pivotal role AI and automation play in shaping the future of DevOps and innovation.

Empowering Through AI and Automation

AI and automation are not just buzzwords; they are essential tools that can streamline operations, enhance productivity and drive innovation. Companies specializing in software development for startups, including ideation, MVP development and scaling, illustrate how AI and automation can transform the development process. These technologies enable startups to accelerate their growth and improve their offerings, making them more competitive in the market.

The integration of AI into software development processes can automate repetitive tasks, allowing developers to focus on more complex and creative aspects of their work. For example, AI can automate code reviews, bug detection and testing, significantly reducing the time and effort required for these activities. This not only speeds up development cycles but also improves the quality of the software by catching issues that might be missed by human developers.

Moreover, AI-driven tools can provide valuable insights into user behavior and preferences, helping startups to create more personalized and engaging products. By analyzing large datasets, AI can identify patterns and trends that can inform product design and marketing strategies. This level of insight is crucial for startups looking to differentiate themselves in a crowded market and build products that truly resonate with their target audience.

The DevOps Landscape

In the realm of DevOps, AI is making significant strides. DevOps is a broad and complex field, requiring a blend of development, security, compliance and operational skills. AI can enhance DevOps by optimizing labor-intensive tasks and creating intelligent workflows. While the journey to fully automated DevOps is ongoing, current successes are primarily in monitoring and anomaly detection. AI’s ability to analyze vast amounts of data and identify patterns can lead to more efficient and effective operations.

For instance, AI can be used to monitor system performance in real-time, identifying anomalies and potential issues before they escalate into critical problems. Machine learning algorithms can analyze logs and metrics from various sources, detecting patterns that indicate system health or predict failures. This proactive approach allows DevOps teams to address issues more quickly and reduce downtime, improving the overall reliability of the systems they manage.

AI can also play a crucial role in optimizing infrastructure management. By analyzing usage patterns and performance data, AI can help in scaling resources dynamically to meet demand, thereby optimizing cost and performance. Predictive analytics can forecast future infrastructure needs based on historical data, enabling better planning and resource allocation. This not only ensures that systems run smoothly but also helps in avoiding over-provisioning and under-utilization of resources.

Security and Fraud Prevention

One of the critical applications of AI is in the area of security and fraud prevention. In the realm of e-commerce, AI plays a crucial role in detecting and preventing fraudulent activities. By analyzing patterns and behaviors, AI can identify anomalies and mitigate risks, ensuring a secure and trustworthy platform for users. This capability is essential for maintaining customer trust and protecting company assets.

In security, AI can be leveraged for threat detection and response. Traditional security measures often rely on predefined rules and signatures to identify threats, but AI can go beyond these limitations by using machine learning to detect novel threats. For example, AI can analyze network traffic and user behavior to identify unusual patterns that may indicate a security breach. Once a threat is detected, automated response systems can take immediate action to contain the threat, minimizing potential damage.

In fraud prevention, AI can analyze transaction data in real-time to detect suspicious activities. Machine learning models can be trained on historical fraud data to recognize patterns associated with fraudulent transactions. When a potentially fraudulent transaction is detected, the system can trigger additional verification steps or flag the transaction for further investigation. This proactive approach helps in reducing the incidence of fraud and protecting both businesses and consumers.

Compliance and Regulation

For startups, navigating compliance and regulation is a significant challenge, especially in highly regulated industries like finance and healthcare. Leveraging AI can help streamline these processes, ensuring that startups meet necessary regulatory requirements. By integrating AI into their compliance strategies, startups can build more robust, compliant solutions that are ready for market entry. This proactive approach not only mitigates risk but also positions startups as trustworthy and reliable partners.

AI can assist in automating compliance checks and audits, reducing the manual effort required to ensure adherence to regulations. Natural language processing (NLP) can be used to analyze regulatory documents and extract relevant compliance requirements, which can then be applied to internal processes and systems. Additionally, AI-powered tools can continuously monitor compliance status, providing real-time alerts and reports to ensure that any deviations are quickly addressed.

In the context of data privacy regulations such as GDPR and CCPA, AI can help manage and protect sensitive data. AI-driven data anonymization techniques can ensure that personal information is protected while still allowing data to be used for analysis. Automated data governance tools can track data lineage, ensuring that data usage complies with regulatory requirements. This helps in building trust with customers and regulatory bodies, demonstrating a commitment to data privacy and security.

AWS: A Catalyst for Innovation

Amazon Web Services (AWS) plays a pivotal role in democratizing AI and supporting startups. AWS offers a range of resources, from credits and accelerators to compliance partners, helping startups navigate regulatory requirements and leverage AI effectively. The availability of these resources makes it easier for startups to experiment with and integrate AI into their products and services, accelerating their path to innovation.

AWS provides a suite of AI and machine learning services that startups can use to build and deploy AI-powered applications. AWS SageMaker, for example, allows developers to build, train and deploy machine learning models at scale. With AWS Lambda, startups can run code without provisioning or managing servers, enabling them to execute AI-driven functions in response to specific triggers. These tools, combined with AWS’s extensive documentation and support, provide a robust framework for startups to innovate and grow.

The Path Forward

As we embrace AI and automation, the focus should remain on creating value and solving real-world problems. The journey is not without challenges, but with resilience and innovation, startups can harness the power of AI to drive progress and transformation. It’s essential for startups to not only adopt these technologies but also to use them strategically to create sustainable and impactful solutions.

AI and automation are set to revolutionize the future of innovation and DevOps. By leveraging these technologies, startups can streamline their operations, enhance their productivity and ultimately drive economic growth. The key lies in integrating AI in ways that are meaningful, differentiated and compliant, ensuring that the solutions they offer are both innovative and reliable. As we move forward, the potential for AI and automation to transform industries is immense, and the opportunities for startups to lead this change are unparalleled.

June 30, 2025
The Self-Driving Help Desk: Agentic AI’s Role in the Next DevOps Era
The concept of agentic AI, where intelligent agents perform tasks, not just answer questions, is finally arriving in DevOps and IT. It is an urgent need in the world of DevOps, where teams are increasingly strained by scale, complexity and security demands. The application of AI has added unprecedented velocity to software development that now requires applications to be launched into production at an equitable speed, thus adding further strain on infrastructure teams.

While AI has transformed industries from content creation to autonomous vehicles, most enterprise IT systems still largely rely on static automation scripts and human-run help desks. The agentic help desk model introduces a transformative shift: DevOps engineers and IT administrators are no longer scripting repetitive tasks — now, they can build AI agents that run these workflows in real time enabling an unprecedented level of self-service for end users.

The Challenge With AI in IT Operations

In its current state, AI in IT is largely limited to chatbot-style Q&A interfaces. These tools can provide useful information or even generate snippets of code, but they fall short when it comes to taking action. Critical operational tasks — especially those that involve “write” actions like deploying infrastructure or changing configurations — are typically outside the scope of today’s AI implementations.

This isn’t due to a limitation in language models; it’s a safety concern and also requires a specialized middle layer of orchestration, above the models.

This middle layer — a framework that supports five key components:
1. Context Awareness:
AI must understand the environment in which it operates. For instance, when an engineer asks, “Why is my cart service crashing?”, the AI needs access to metadata like cluster names, deployment status and error logs. Injecting this system context into prompts dramatically improves the relevance and utility of responses.
1. Centralized Tools and APIs:
In large-scale DevOps environments, agents need to invoke APIs for systems like AWS, Kubernetes, or Terraform. Relying on every agent to independently implement these APIs is inefficient. A centralized, standardized set of tools — delivered via a shared Model Context Protocol (MCP) — allows for more scalable and secure automation.
1. Apply Guardrails with Human-in-the-Loop Capabilities:
Autonomy shouldn’t mean loss of control. AI systems should include approval mechanisms, letting engineers review or augment workflows before they are executed, as well as collaborate across teams. Engineers need peer-programming-like experiences that support terminals, pull requests, and inline code suggestions within CI/CD workflows. This is essential for safety and trust, and mirrors how an engineer collaborates with a teammate.
1. Granular Access Control:
Access should inherit user permissions and be enforced through a secure framework. Without this, organizations risk accidental privilege escalation or security breaches. Agents should also operate entirely within the customer’s cloud — no remote control plane — preserving security, data locality, and compliance, with strict RBAC, audit trails and scoped, time-bound credentials.

Why a Help Desk Interface?

The traditional help desk is a familiar construct in IT: Users submit tickets, and human specialists resolve them asynchronously. This system is predictable — but it’s also slow and inefficient in a world where things are moving fast and surrounded by automation.

The agentic help desk reimagines this model. Instead of waiting in a queue, users assign tickets to intelligent agents capable of resolving them in real-time. Administrators design, curate and approve agents to handle specific workflows, and end users simply describe their needs in plain language. Agents can even transfer tickets to one another based on specialization.

This evolution supports a future where support is continuous, real-time, and intelligent — without sacrificing safety or control.

Real-World Use Cases

Agentic workflows are already proving valuable in high-pressure DevOps environments. Common examples include:
- Translating legacy deployment formats (e.g., Docker Compose) into Kubernetes-ready manifests
- Running cloud cost optimization diagnostics and implementing savings suggestions
- Troubleshooting performance issues using observability frameworks like OpenTelemetry
- Onboarding new engineers with personalized, interactive guidance
- Automatically detecting and remediating security policy violations
By operationalizing tasks with AI agents, teams are able to reduce manual tasks, accelerate deployments and free up senior engineers to focus on strategic priorities.

The Road Ahead: From DevOps to Broad IT Support

While the initial focus is on DevOps, the underlying architecture of agentic help desks is broadly applicable across the IT ecosystem. Any function that currently relies on ticket queues — security operations, IT support, compliance audits — can benefit from real-time AI-driven workflows.

The long-term vision is for software vendors and in-house teams to build and publish their own agents, creating a dynamic network of automation. This model is extensible, customizable, and built for a future where scale, speed and compliance are all critical.

As IT systems grow more complex, the reactive help desk model is no longer sustainable. Agentic AI introduces a paradigm shift — one where automation becomes intelligent, collaborative and proactive. The potential for this technology to enhance velocity, reduce burnout and harden security is enormous.

But this future won’t be realized through LLMs alone. It will require a deep rethinking of infrastructure, interfaces and governance — built not just for intelligence, but for impact.
June 30, 2025
Google Adds Gemini CLI to AI Coding Portfolio

The command line interface has long been the developer’s sanctuary, a place where efficiency meets simplicity, and where the most complex tasks can be distilled into elegant commands. Now, Google is enhancing this sacred space with the Gemini CLI. This open-source AI agent transforms how developers interact with their terminals while maintaining the lightweight, powerful experience they love.

Breaking Down Barriers to AI-Powered Development

Traditional AI coding assistants often require switching contexts, opening new applications, or navigating complex interfaces. Gemini CLI eliminates these friction points by bringing Gemini 2.5 Pro directly to where developers already spend most of their time. This isn’t just another AI tool; it’s a fundamental reimagining of how artificial intelligence can seamlessly integrate into existing workflows.

The beauty of Gemini CLI lies in its approach to solving a common developer pain point: The constant context switching between development environments and AI assistance. Instead of disrupting established workflows, it enhances them by providing intelligent assistance exactly where and when it’s needed.

Accessibility for Individual Developers

The most compelling aspect of Gemini CLI is its generous usage model. With just a personal Google account, developers gain access to Gemini 2.5 Pro with an impressive 1 million token context windows. The service offers 60 model requests per minute and 1,000 requests per day at no charge, easily the most generous free tier in the industry.

This approach democratizes access to cutting-edge AI capabilities, ensuring that individual developers, students and small teams aren’t left behind in the AI revolution. Whether you’re debugging a complex issue at 2 AM or exploring a new framework during your weekend projects, Gemini CLI provides the assistance you need without worrying about usage limits or costs.

Beyond Simple Code Generation

While many AI coding tools focus primarily on code completion and generation, Gemini CLI takes a more holistic approach to developer productivity. Its built-in capabilities extend far beyond writing code:

Real-time Research Integration: The tool can ground prompts with Google Search, fetching web pages and providing current, external context. This means you can ask questions about the latest framework updates, security vulnerabilities, or best practices and get accurate, up-to-date information without leaving your terminal.

Dynamic Problem Solving: Rather than providing static responses, Gemini CLI can engage in iterative problem-solving sessions, helping you work through complex debugging scenarios or architecture decisions step-by-step.

Workflow Automation: The tool can be invoked non-interactively within scripts, enabling automation of repetitive tasks and integration with existing CI/CD pipelines.

File and Command Manipulation: Unlike web-based AI assistants, Gemini CLI can directly interact with your local file system and execute commands, making it a proper terminal-native solution.

Open Source Foundation Drives Innovation

Google’s decision to release Gemini CLI under the Apache 2.0 license reflects a commitment to transparency and community-driven development. This open-source approach serves multiple purposes: it allows developers to inspect the code for security implications, understand exactly how the tool works, and contribute improvements back to the community.

The extensibility built into Gemini CLI is particularly noteworthy. Support for the Model Context Protocol (MCP) and customizable system prompts through GEMINI.md files mean developers can tailor the tool to their specific needs and workflows. This flexibility ensures that Gemini CLI can adapt to diverse development environments rather than forcing developers to adapt to it.

Integration With Existing Ecosystems

The integration between Gemini CLI and Gemini Code Assist demonstrates thoughtful ecosystem design. Developers can seamlessly move between terminal-based workflows and IDE-based development, maintaining consistent AI assistance across different contexts. This unified approach means learning one tool translates to productivity gains across multiple environments.

In VS Code, the agent mode capabilities mirror what’s available in the terminal, providing multistep planning, auto-recovery from failed implementations and collaborative reasoning. This consistency reduces cognitive load, allowing developers to focus on solving problems rather than learning different interfaces.

“There’s no more fundamental tool than the terminal for software developers, platform engineers, SREs, and software testers. Seamlessly integrating Gemini CLI into the terminal window means users of all types can fluidly move between IDE, terminal, configuration management, infrastructure-as-code and many other tools,” said Mitch Ashley, VP Practice Lead of Software Lifecycle Engineering at The Futurum Group. “I predict we’ll see a great deal more beyond software coding happening with Gemini CLI. The terminal equipped with the same AI used for coding tasks can automate and orchestrate workflows, tasks, and system administration functions across the software development lifecycle. You can only imagine what a platform engineer focusing on improving developer productivity might do with MCP servers, 60 model requests per minute, and 1,000 requests per day for free.”

The Future of Developer Productivity

Gemini CLI represents more than just another AI tool; it’s a glimpse into the future of developer productivity. By respecting developers’ existing workflows while enhancing them with intelligent assistance, it demonstrates how AI can truly augment human capabilities rather than replace them.

The tool’s emphasis on extensibility and community contribution suggests a future where AI assistance becomes increasingly personalized and tailored to individual needs. As the open-source community contributes extensions, plugins and improvements, Gemini CLI will evolve to serve an ever-wider range of use cases and development scenarios.

For DevOps professionals, this means AI-powered assistance for infrastructure management, deployment automation and troubleshooting complex distributed systems, all accessible through the familiar comfort of the command line.

Getting started with Gemini CLI requires only an email address and a few minutes of setup time. In an industry where productivity tools often come with steep learning curves and hefty price tags, this accessibility is refreshing and powerful.

The terminal has always been about efficiency and control. Gemini CLI upholds these principles while introducing the intelligence necessary to address tomorrow’s development challenges.

June 30, 2025
“DevOps is Dead? Long Live DevOps-Powered Platforms”
I knew PlatformCon 2025’s first New York City Live Day would bring the usual debates about whether platform engineering is the new DevOps. What I didn’t expect was a guy roaming the lobby in a white sash screaming, in block-letter embroidery, “Platform Engineering is Dead.” If the goal was to provoke, mission accomplished — but the irony was thick. You could barely move in the sold-out Convene space, shoulder-to-shoulder with more than 400 practitioners eager to compare Terraform modules, golden paths and OpenTelemetry pipelines.

PlatformCon’s final registration numbers — 40,000+ virtual sign-ups and a projected 100,000 session views — tell a different story: Platform engineering is very much alive. And the biggest takeaway for me, a DevOps believer, is that the rivalry between DevOps and platform engineering has quietly faded. Instead, we’re watching an overdue reconciliation that promises to make every release pipeline faster, safer and a lot more fun.

From Turf War to Tag Team

Back in 2022, you could scarcely open LinkedIn without a hot take declaring “Platform Engineering will replace DevOps.” The framing never made sense —DevOps is a culture and set of practices; platform engineering is an organizational response to scale. That distinction finally felt settled in NYC. Speaker after speaker pointed out that a healthy Internal Developer Platform (IDP) is a force multiplier for DevOps teams, not a replacement.

Kelsey Hightower’s morning keynote set the tone. “Platforms aren’t magic APIs,” he reminded us. “They’re agreements that make engineers faster at delivering business value.” Our job in DevOps has always been to accelerate that value stream. The platform engineer just gives us better raw material to work with — golden-path templates, standardized environments and opinionated guardrails. Less yak-shaving, more shipping.

AI: The Great Convergence Engine

If DevOps and platform engineering needed a common enemy — or ally — to bond over, AI provided it. A panel featuring Nvidia, Google, Rootly and Thoughtworks explained how large language models are automating “the last mile” of toil, from incident response bots that reason over Grafana dashboards to code-gen pipelines that spit out compliant Terraform.

Workshops went even deeper. Gitpod walked attendees through an “AI Technology Radar” exercise to decide which GenAI tools to adopt, trial, assess or hold. And then there was Sedai’s bravely named “Self-Driving Cloud” lab: Think agentic AI closing the feedback loop between cost, performance and deployment decisions with almost no human in the path.

For DevOps engineers, this means two things:
1. The cognitive load shifts. Troubleshooting a flaky deployment may start with prompting an LLM that already read all your runbooks.
2. The platform boundary blurs. If AI can orchestrate infra and app layers together, DevOps and platform teams must design those control planes side by side or risk “dual brains” fighting for supremacy.
Observability: The New Gravity Well

Every conversation about platform maturity eventually landed on observability. Mezmo’s hands-on OTel pipeline workshop was standing-room only; folks were literally sitting on the hallway carpet to follow along. The logic is straightforward: You can’t automate what you can’t see. For DevOps practitioners, high-fidelity telemetry is now table stakes — whether you’re feeding an agentic AI, debugging an ephemeral sandbox, or proving compliance to auditors. Expect platform blueprints to ship with observability baked in, not bolted on.

Sponsors Tell the Story

Look at the badges behind every coffee urn and you’ll spot familiar DevOps and DevSecOps logos — GitHub Actions, Mezmo, Teleport, Cortex, Sedai, Tailscale. Many of these vendors cut their teeth in CI/CD, IaC, or shift-left security long before “platform engineering” was a LinkedIn hashtag. Their presence reinforces a simple point: The DevOps ecosystem isn’t being cannibalized; it’s expanding into platform architecture, policy-as-code and developer portals.

Three Big DevOps Takeaways
1. Platform = Product = Velocity
  
  Treating the platform like a product — complete with backlogs, SLAs and NPS — gives DevOps teams the reliable substrate they’ve always begged for. Less firefighting frees us to optimize flow, not fix YAML.
2. Agentic AI Will Test Your Guardrails
  
  Auto-remediating bots and self-driving clouds are amazing until they deploy a breaking change at 2 a.m. DevOps pros must harden policy engines, progressive delivery gates and observability loops to keep the machines honest.
3. Collaboration Over Handoffs
  
  The healthiest orgs I interviewed position DevOps inside the platform team or vice versa. That unity prevents the “throw it over the wall to the platform squad” anti-pattern and keeps integration and deployment from becoming the next bottleneck.
About That “Platform Engineering is Dead” Sash

So why the funeral garb? My guess: A tongue-in-cheek jab at hype cycles. Just as “DevOps is dead” clickbait pushed us to sharpen our message, the sash was a reminder that real value — not buzzwords — keeps a movement alive. Judging by the hallway traffic and workshop queues, platform engineering is passing that test.

Where We Go Next

I left Midtown convinced that we’re entering a post-label era, where the strict DevOps vs. platform binary distinction dissolves. High-scale organizations already blur the lines: Platform engineers own golden paths; DevOps engineers extend them; both instrument the heck out of everything, all while teaching an LLM to babysit the cluster at 3 a.m.

Expect to see:
- Shared scorecards that measure both platform adoption and deployment frequency.
- Joint SLO ownership where DevOps and platform teams co-sign error budgets.
- AI-native incident command rooms where bots write the first draft of every post-mortem.
Final Word

DevOps isn’t dead, and neither is platform engineering. The sash guy had it backwards: The rivalry is what’s dead. In its place, we have a pragmatic partnership aimed at giving developers the safest, fastest runway imaginable. That’s a win for everyone — from the two-pizza team pushing to prod on Friday afternoon to the CISO sleeping better at night.

If PlatformCon NYC proved anything, it’s that the DevOps ethos of collaboration, automation and continuous improvement now has a bigger, better stage. And I, for one, can’t wait to see what happens when the lights come up and the AI co-pilots take the wheel.
June 28, 2025
Secure Code Warrior Defines Security Rules for AI Coding

Secure Code Warrior has made available a set of security rules for application developers using artificial intelligence (AI) tools to generate code.

Company CTO Matias Madou said the AI Security Rules made available on GitHub are intended to encourage developers to review code generated by AI for security issues that these tools may inadvertently introduce. Most AI tools were trained using samples of code that were randomly collected from across the web. As such, they can introduce vulnerabilities simply because they were trained using code that contains similar flaws, noted Madou.

Additionally, the AI Security Rules encourage developers to establish guardrails that steer AI away from risky patterns and common security missteps, such as insecure authentication flows or failure to use parameterized queries.

Previously, Secure Code Warrior has made available AI security training courses for application developers that Madou said will continue to be expanded.

In general, AI coding tools have, from a cybersecurity perspective, plusses and minuses. AI coding tools are, for example, unlikely to introduce a SQL Injection vulnerability. In fact, there may come a day when this particular vulnerability is completely eliminated as AI coding tools become more widely used, noted Madou. However, AI coding tools also introduce other potential security concerns, including hallucinations.

It’s not clear to what degree developers are relying on AI tools to generate code, but a recent Futurum Group survey finds that 41% of respondents expect generative AI tools and platforms will be used to generate, review and test code.

Additionally, The Futurum Group finds organizations are exploring multiple paths to employing artificial intelligence (AI) capabilities across the software development lifecycle (SDLC). Over the next 12 to 18 months organizations plan to increase spending on not only AI code generation (83%) and agentic AI technologies (76%), but also existing familiar tools that have been augmented with AI.

In effect, there is now a battle underway between incumbent providers of tools and startup rivals for the hearts and minds of software engineers.

Regardless of how code is written, an application developer remains accountable for its quality. The challenge is that over time there is often a tendency to trust the output of an AI coding tool too much. Inevitably, the code generated might not only contain vulnerabilities but might also be deeply flawed in ways that might prevent an application from running at all. Just as challenging, AI coding tools might not be of much help when updating existing application environments simply because they might not have enough context to generate code that will work, noted Madou.

On the other hand, there are many developers that only have a rudimentary appreciation for security issues. AI coding tools in those instances can significantly improve the quality of the code that might otherwise have been created by an application developer.

There is, of course, no going back. AI coding tools are here to stay. The real issue is putting the guardrails in place that minimize any security vulnerabilities or other types of flaws that might compromise the security of an application.

June 26, 2025