What are the main challenges you are facing in the project?

The primary challenges in our current project include data consistency across distributed services, scaling to handle peak traffic, and ensuring low latency for real-time analytics. Data consistency is difficult because we use a mix of relational and NoSQL databases, leading to eventual consistency issues that can surface during high concurrency. Scaling requires dynamic resource allocation; we rely on Kubernetes autoscaling, but we still face cold-start delays for stateless functions. Latency is a concern for our real-time dashboards; we need to process millions of events per second while keeping end-to-end latency below 200 ms. Additionally, integrating third-party APIs introduces unpredictable network latency and versioning conflicts. To mitigate these, we are adopting event sourcing, implementing circuit breakers, and using a CDN for static assets. Continuous monitoring and automated rollback strategies are also part of our mitigation plan.