Hello everyone. I'm excited to be here at Con 42 SRA 2025. My name is Villa Kumar here and I'm senior cloud platform developer with over 20 years of experience. Today I will share cloud native reliability strategies that blends automation, multi-region, archite, and ai. All aimed at minimizing downtime and maximizing confidence in your platforms. Cloud native reliability strategies. So today we will explore how cloud native environments can be designed to maximize uptime, umbrella automation, and integration with AI tools to prevent outages before they occur. Whether you're running microservices, Kubernetes cluster or multi-region infrastructure, this talk will arm you with best practices and strategies. Why reliability matters. Downtime is more than a technical issue. It's a business risk. It directly impacts the revenue, customer satisfaction, and brand trust. In a cloud native world where systems are always on reliability must be intentional and proactive. It's not enough to react to the incidents. We must prevent them, detect them early and recover quickly. That starts with treating reliability as a core architectural goal. So what are the key cloud reliability challenges? Cloud native systems bring agility, but also complexity. You have likely faced latency spikes from autoscaling, case scaling failures between services and the fatigue of noisy alerts. These challenges are amplified by dynamic environments and ephemeral infrastructure. The key is visibility, knowing what happening in your en, in your environment. When and why And creating system that can withstand this unpredictability. So multi legion failover one pro and reliability strategy is geographical redundancy. Deploying your application in multiple regions. Protect against zone outages, network dis disruptions. Or cloud provider failures, but redundancy is not enough. You must test it. Regular failover rails, kiosk engineering and synthetic traffic can validate your setup before a real incident occur. Active models give the best availability, but that requires more engineering effort. So self-healing automation, we all know that humans don't scale. Self-healing systems are your best defense against unexpected disruptions. With Kubernetes, you can autostart containers. With tar Terraform, you can rebuild infrastructure. Declaratively Ansible can automate your incident responses. The goal is to detect, react, and recover, ideally without a pager. Alerting a human at 3:00 AM so AI driven reliability AI is no longer futuristic. It's here and helping us keep systems online. AI model can forecast anomalies, reduce alert fatigues, and with intelligent filtering. Identify root cause faster than we think. Platforms like New Relic, ai Dyna, and Azure Monitor are already embedded in enterprise ob observability tracks. This isn't about replacing engineers, it's about giving them superpower. So what are the best practices? Here is the checklist designed for failure. Simulated with chaos tools, automate deployments, observe everything, logs, metrics, traces, and real time user data. Promote share responsibility. Reliability is not just the SRE job. Define the tracks. SLOs, SLI and other budgets adopt infrastructure as a code and ops workflow. So what are the final takeaways? If there is one thing I hope you live with, it's resilience isn't just about preventing downtime, it's about earning users trust. Cloud native reliability is a continuous journey for a better design, smarter automation and collaborative culture. Let's build system that predict failures and keep your environment up and running with minimal downtime. Thank you. I would like to hear your questions thought. Or challenges you face, please reach out to me or an email or LinkedIn. Thank you.