I am expanding our internal research initiative, NetMetrics, and need a streamlined way to gather and analyse connection quality from several geographic locations. Your task is to design and hand over a working solution that: • Deploys lightweight, authorised agents on Windows, Linux, macOS, and Android machines already in place. • Continuously measures latency, uptime, and packet-loss between those agents and a central endpoint, then securely ships the data back. • Stores the results in a format that supports three parallel outputs: – real-time or near-real-time graphs and charts, – raw data logs for deep dives, – concise summary reports our engineers can skim. Core expectations 1. Agent scripts or binaries must be easy to install, self-healing on reboot, and configurable for test frequency. 2. All traffic stays private—no public VPNs, proxies, or third-party relays. Please rely only on standard, widely audited libraries and protocols (e.g., ICMP, TCP, or HTTP pings; secure REST or gRPC). 3. Provide clear documentation: deployment steps, config samples, and a quick guide to extending tests to new locations. 4. Include a modest visual dashboard (Grafana, Kibana, or similar) that we can host ourselves for live views of the collected metrics. 5. Deliver unit-testable code and a short validation checklist so we can confirm readings match expected round-trip times on our QA servers. If you have prior experience with distributed monitoring frameworks (Prometheus/node-exporter, Telegraf + InfluxDB, Zabbix, or custom Go/Python stacks), mention it—those skills map perfectly to this job.