Hello DEV! SRE here..👋

Hello, DEV 👋 After spending years working in Site Reliability and Platform Engineering, I figured it was finally time to start writing about the things I spend most of my days thinking about. I am a Staff Site Reliability Engineer based in Canada, focused on building and scaling reliability for large-scale payment infrastructure. My world revolves around things like: ⚙️ Distributed systems 🛡️ Reliability and resilience 🔄 High availability and failure recovery 🔍 Observability 🔐 Transactional integrity and idempotency 🚨 Incident response Basically, figuring out what happens when things inevitably break and designing systems so that failure doesn't turn into an outage. I didn't start out working on financial systems. I came through platform and reliability engineering the less glamorous way, including automating infrastructure for a provincial energy regulator before moving into increasingly complex enterprise systems. Along the way, I've learned that some of the most interesting engineering problems aren't about making systems work. They're about making them keep working when everything around them doesn't. What I'll Be Writing About ✍️ I created this account to share some of the things I've learned, experimented with, and occasionally gotten spectacularly wrong. Expect a mix of: Distributed Systems Architecture patterns, concurrency, state management, failure modes, idempotency, messaging, and the trade-offs that don't usually make it into architecture diagrams. SRE & Observability Incident response, alerting, telemetry, SLOs, debugging production systems, and the operational problems that look simple until you're the person on call. AI × Reliability 🤖 I'm particularly interested in where agentic AI meets production engineering. Not just "what can an LLM do?" But: How do you build AI systems that behave reasonably when networks fail, workers crash, messages are duplicated, and the model itself gets things wrong? That's where things get interesting. I'll be using DEV to document the experiments, architecture decisions, failures, lessons learned, and the occasional rabbit hole along the way. If you're interested in SRE, distributed systems, observability, or production AI, stick around. And if you're already working in these areas, I'd love to learn from you too. Find me here 🔗 💼 LinkedIn: https://www.linkedin.com/in/kashyapkohli 💻 GitHub: https://github.com/k-kohli10 What's something you've learned the hard way while building or operating production systems? 👇