Worked on enhancing performance and reliability in the grafana/prometheus repository by implementing a read-concurrency improvement for scrape targets. Addressed contention issues in read-heavy environments by refactoring the DiscoveredLabels and URL components to use RWLock, specifically RLock, instead of a standard Lock. This approach enabled concurrent reads, boosting throughput for metric collection under high load and supporting high-cardinality targets more efficiently. The work focused on concurrency and performance optimization using Go, with changes traceable through a dedicated commit. No bugs were fixed during this period, and the contribution centered on a single, targeted feature to improve system scalability.