Platform engineering is about building platforms that support teams in developing and deploying software quickly and reliably. It involves providing standardized tools and workflows to make the day-to-day of building applications easier. Platform engineers focus on building "paved roads" for their teams, which are clear, standardized paths for building and deploying software, reducing cognitive load without taking away autonomy. Observability is a crucial part of platform engineering, as it provides timely feedback on performance, errors, and user engagement to improve release quality, increase release frequency, and operational efficiency. Platform engineers need to build observability solutions that minimize cognitive load and tool sprawl, are fast and easy to implement, easily customizable, scalable, and offer features such as alert notifications, log management, application performance monitoring, distributed traces, Kubernetes monitoring, real user and synthetic monitoring, infrastructure and network monitoring. To implement observability in their platform, platform engineers should aim for 100% observability, consider data security before gathering monitoring data, set up monitoring for essential metrics, deploy applications with APM, include other services that applications rely on, automate everything, including default alerts and dashboards, and use observability as code.