Project Name : Network Log & Telemetry Pipeline (Syslog → Parse → Alerts)

“I implemented a lightweight log pipeline that ingests syslog, parses events, detects anomalies, and raises alerts.”

Features  UDP syslog receiver (lab scope) → writes to SQLite/Postgres.  Parsers for common patterns (link up/down, OSPF neighbor, CPU high).  Rules & thresholds: burst of link flaps, error rate spikes.  Alerts via email/Teams/Slack webhook; daily digest report (Pandas).  Export dataset for Power BI dashboard or Grafana JSON.

Folder Structure network-log-pipeline/ ├── main.py # Entry point: starts syslog receiver and scheduler ├── parsers.py # Syslog parsing logic ├── alerts.py # Email/Slack/Teams alerts ├── database.py # SQLite/Postgres connection & models ├── requirements.txt # Libraries ├── config.yaml # Ports, thresholds, webhook URLs ├── sample_data/ # Sample syslog text files for testing └── reports/ # Daily digest CSV/HTML

Descirption : “One of the projects I worked on was building a lightweight network log and telemetry pipeline. The main challenge we had was that network devices were sending syslog messages, but there was no centralized way to parse them, store them, or generate alerts for unusual activity. Engineers had to manually look at raw logs, which was not scalable. My task was to create an automated pipeline that could ingest syslog events in real time, parse them into a structured format, and trigger alerts when anomalies or critical events occurred. To implement this, I developed the pipeline in Python. I set up a UDP listener to receive syslog messages from devices, then used custom parsers and regular expressions to extract useful fields like timestamp, device, and event type. These logs were stored in a structured database for easy querying. I also integrated FastAPI to build a simple web interface so engineers could search and filter logs. On top of that, I built an alerting mechanism that flagged unusual patterns, such as repeated link-down events, and sent notifications to the team. As a result, log analysis became much faster, and the team could identify issues proactively instead of waiting for outages. The pipeline also reduced manual monitoring efforts and improved incident response time because alerts were generated in near real time. Overall, this project gave me strong experience in Python automation, parsing real-time data, and building monitoring tools that directly improved network reliability.”