Python System Monitoring Automation

I’m looking for a Python-savvy coder who can craft a lightweight automation script focused on system monitoring and the data processing that comes with it. The idea is simple: collect real-time metrics from the host machine, process that raw data into something meaningful (think rolling averages, trend detection, outlier flags), and store or forward the results so they’re easy to consume.

Here’s what I need the finished script to handle:
• Poll key resources—CPU, memory, disk, network, or any other common metric you recommend—at an adjustable interval.
• Process each batch of readings on the fly rather than dumping unfiltered logs; basic stats plus optional alert thresholds are fine for now.
• Output the cleaned data to either a local file (CSV/JSON) or a lightweight endpoint such as an SQLite DB or REST API—whichever you feel is most maintainable.
• Keep dependencies minimal: standard library wherever possible, otherwise well-supported packages like psutil or pandas.
• Clear, inline comments and a short README so I can tweak intervals, thresholds, or output paths without digging through code.

Acceptance is straightforward: I’ll run the script on a Linux box, let it collect for 15 minutes, then verify that the processed data matches the live metrics and that alerts (if any) trigger correctly.

If you’ve built similar monitoring or data-processing automations before, feel free to mention the tools you used or suggest small improvements that would make this more robust.