WaelDaagi

Built a secure delivery platform on Kubernetes, from the first commit scan to runtime detection.

• Built a reusable 7-stage GitLab CI/CD component (secret scanning, dependency and container vulnerability scanning, image signing, Kubernetes signature verification) adopted by two projects; onboarding now takes six lines of YAML. Its first run caught a live AWS access key hidden in Git history.
• Migrated a NestJS and MongoDB service to production Kubernetes, fixing 4 security defects including root container execution and 2 critical CVEs (RCE, XSS); hardened the Dockerfile (npm ci, non-root user, tini).
• Delivered a production-ready Kubernetes platform in 9 phases, verified by 65 automated conformance checks, with Argo CD GitOps delivery; cut deployment latency from about 3 minutes to seconds with event-driven Argo Events.
• Configured merge-request pipelines with automated tests and security scans; debugged and fixed Docker-in-Docker TLS failures on self-hosted GitLab Runners (Linux, systemd).
• Set up Prometheus, Grafana and Loki observability, and wrote the technical documentation, including disaster recovery.

• GitLab CI/CD
• Kubernetes
• Argo CD
• Kyverno
• Trivy
• Docker
• Prometheus
• Grafana

Turned unstructured documents into classified, structured records for a document management system.

• Built a Python data processing pipeline that turns unstructured PDF, DOCX and scanned image files (Tesseract OCR, OpenCV) into classified, structured records for an Electronic Document Management System, with confidence scoring.
• Trained and evaluated a TF-IDF and Multinomial Naive Bayes classifier (scikit-learn, pandas), with batch processing and result monitoring through a Streamlit interface.

• Python
• scikit-learn
• pandas
• Tesseract OCR
• OpenCV
• Streamlit