Hello everyone. This is koala. I'm a senior DevOps engineer working for Rotter. So today my topic is the synergistic impact of AI in DevOps. So today I'm excited to talk about something that is, shaping the future of engineering. The convergence of a and DevOps. As many of DevOps has transformed. How we release software enabling fast iterations, automation, and also a culture of collaboration. But what happens when we inject AI into this already powerful lifecycle? We get intelligent automation that adapts, predicts and loans from data. We move from reactive processes to predictive. And also perspective ones. The shift means, fewer outages, more stability systems, and, faster feedback loops. It's not just about writing code and, shipping fast and fast anymore. It's about shifting shipping. Smart. Let's take about talk about a key takeaways from this project like our which I'm presenting like a adds like a predictive intelligence to automation and also like a DevOps plus ai, which is autonomous and adaptive delivery cycles. It enables smarter decisions across the software lifecycle. Transforms feedback loops into learning systems, turns raw data into actionable operational insights. So let's talk about the ai assisted development tools. AI is now sitting next to developers in the IDs. Tools like GitHub copilot, Amazon Code Whisperer, and TAM nine are revolutionizing how we build software. Let's talk about like how this what these tools do and how they're very useful. They, auto complete full functions and boiler plate logic and also suggest variable names logic improvements and also fixing them generates unit tests and documentation on the fly. And also their adapt to. They adapt the code base and also like a developer style over time. So with ai developers move faster learn quicker, and also produce a cleaner code. No more searching stack workflow for Synex help these tools act extra experienced mentors helping engineers write a better code and also reduce bugs from day two, from day one. And also it the impacts like by using these tools are like, 30 to 50% faster development, speed lower onboarding friction for new development to developers higher consistency and reduce the technical depth. So let's talk about GitHub co-pilot which is a AI coding assistant. GitHub Copilot deserves a special mention because it's not just auto complete. It's contextual code project prediction at scale. Copilot learns from millions of open source projects and adapts to your project in real time. If you want to talk about why Copel stands out, it understands, context, functions and dependencies, and also can build entire block of logic from a common prompt. And also it suggests like a test and its edge cases and, writes talk strings. Imagine let's imagine writing a function. To process a CSV data co-pilot, not only reach the header, but suggest like how to handle errors and, misleading fields. That's next level productivity. As eruption grows, teams are saving hours per week and also increasing our test coverage without trying and additional benefits are like, it enhances paid programming acts as a live documentation. And also boost our creativity and experiment experimentation. Let's talk about a for tive pipeline management. So we all know CAC, right? Until pipelines fail mysteriously a helps us like, prevent them, it is very interesting, like by analyzing build histories and also commit metadatas. And also job patterns air flags issues before they happen. A, in pipeline animals, like a prediction of, flaky test and also like unstable branches, smart reordering of job execution for performance, and also like detection of regressions before matching to Maine dynamic resource location to reduce cost and waste. So while time, these intelligent adjustments make pipelines faster and far more reliable, developers get faster feedback. Least the reli teams get cleaner artifacts and also like businesses get better uptime. Let's talk about the real time world scenarios, like where it is really valuable up to 40% faster pipelines and also 20% fewer fail bills. Like you can have confidence scoring for every commit. So let's talk about a Sigma solve which is a predictive pipeline management. Let's explore like a real time example. Sigma solve this platform adds an a layer on top of your CACD workflow. What Sigma solve does, is it analyzes, commits test coverages and file vol volatility. And also scores each build from zero to a hundred for deployment readiness. And also flags deployment that resemble patched high risk pattern suggests like a rerouting high risk builds to staging r qa. Sigma solve turns your pipeline into a risk aware engine. It also learns from your team's behavior and address. Its threshold. As your code base evolve developers trust it. Managers use it to improve velocity. Let's talk about a team outcomes. It'll be used for fewer rollbacks and more confidence in fast shipping and also a better collaboration between development and the operations. So let's talk about a, in a DevOp security. Security is often the bottleneck in fast delivery. A helps, close the gap between speed and safety security challenges. A, what a s are like, secrets hard coded in your pool request. Ly permissive IAM policies infrastructure drift from Terraform to AWS and also behavior anomalies in service to service calls with a, you can scan for patterns across millions of logs, not just signatures. A learns from breaches. Across the globe and once you, before you make the same mistake, it clusters alerts and reduces false positives and recommends remedy remediations. Bonus features are like NLP modules that automate summarize incident reports risk scoring by repository environment GI hooks that enforce, like a security posture early. Let's talk about a DevOps metrics improvement with a doesn't just help, it shows results. Teams using AI in the DevOps pipeline have seen massive implements in their core metrics. Let's talk about some of the measurable implements deployment frequencies which are increased by twice. Lead time for cha changes 50% reduction hours from hours to minutes. Like change failure rate is reduced by 30%. A helps, teams correlate incidents. To metrics. If a test fails, often a flags it. If a release causes spikes in latency, AA connects it to the commit. Instead of reacting to the dashboards, you respond to insights as observability to action automated RCA like a root cause analysis, and also like a smart alerting that ties, logs to specific deploys. And also it suggests to roll back, scale up or like a change Confis. So let's talk about a financial services a applications in finance every second matters, right? And also does every transaction a helps institutions secure systems ensure uptime and maintain compliance use cases in finance are like, predicting night batch job delays. Detecting out of patent a PA usage flagging unauthorized config changes auto generating audit trail from CACD metadata. With ai finance teams can operate fast without sacrificing security. It's like having a compliance assistant monitoring your pipelines compliance meets velocity transplant traceability. Prior to governance and also like a built-in risk analysis on deployments. So let's talk about e-commerce. AI DevOps applications. E-commerce is a battlefield of, milliseconds, customers don't wait. A ensures the backend scales just as fa fast as user demands key enhancements with ai r like, like eight four. It is used to forecasting traffic spikes for pre-cal cashy preloading based on user trends and also detecting slow queries in the real time rerouting users during edge outage. And also let's imagine like deploying a new card service, like 10 minute before a flash sale. AA monitors latency and the ab bots like, a rollout if users, conversations drop. And also bus let's talk about the business outcomes. It is used to, give zero down time during events like a Black Friday. Better user experience and more conversations. And also faster resolution of slow paths. And the last talk about human AI collaboration in DevOps. AI won't replace, like a DevOps engineers like, it'll make them temps more efficient and effective. A automates like a test generation alert, prioritization, documentation, and the summaries. Humans still handle them by like designing resilient architectures, navigation, navigating trade offs, and also like business logic crisis management and communication. Together, they create a more scalable. Efficient and human centered development culture. AI is your partner, like not your replacement. Let's talk about the future prospects for AI in DevOps. We are just casting started. Like the feature of a DevOps includes conversational DevOps what's my current release statuses and also like auto. Generating run books and dashboards. S that understand your observability platforms and also autonomous pipelines that self tune based on like a cost load. As LMS improves we will build like a system that don't just react. They all sell off and also by 2025 we have we already saw like a lot of aa involvement happening. Looking ahead I think we use like more sophisticated a models like where that understand more and generate better code across like multiple programming languages. Yeah. So thank you guys. Hope this p pity helps for you like to understand how AI is reshaping DevOps. Not by replacing what we built, but by enhancing it. Let's use this momentum to build smarter, more reliable and more human centered know, like systems. So the future of DevOps is intelligent and, we will be building for it. Thank you.