Specializing in observability, automation, and self-healing infrastructure. Transforming complex distributed systems into highly available, scalable solutions.
Building resilient, scalable infrastructure with cutting-edge technologies
Authoring structured task specifications and evaluation rubrics for assessing LLM performance on infrastructure and cloud engineering scenarios. Implementing reference solutions and automated test suites ensuring task reproducibility at scale across AI evaluation pipelines. Building containerized evaluation environments with Docker for large-scale benchmarking runs and collaborating with AI research teams to iterate on benchmark design and reliability metrics.
Architected a cloud-native security scanning platform on GCP using a microservices design pattern, enabling real-time automated scanning for OWASP Top 10 vulnerabilities and infrastructure misconfigurations. Designed end-to-end CI/CD pipelines with GitHub Actions and Terraform for zero-downtime releases. Developed Python automation tools to auto-generate compliance reports and applied cloud security best practices including IAM role separation and VPC segmentation.
Delivered end-to-end consulting engagements in network auditing, DevOps implementation, cloud migration strategy, and infrastructure security for clients in telecom, finance, and technology sectors. Led cloud migration projects from on-premises to GCP and AWS, improving performance and reducing costs. Enhanced observability through integrated Grafana dashboards for proactive issue detection and reduced MTTD across client environments.
Managed product lifecycle and technical operations for Samsung and as PM of Knox enterprise security platform. Directed digital transformation initiatives, increasing customer engagement and online sales. Achieved notable retention and portfolio growth metrics in a highly competitive and closed local market, negotiated partnerships, optimized product offerings, and improved sales operations efficiency.
Managed secure telecom infrastructure for classified financial investigations. Improved network monitoring and data control systems across regulatory units.
Led audits of large-scale IT infrastructure and telecommunications systems. Provided technical recommendations to enhance system uptime and security compliance.
Cutting-edge solutions in SRE, Cloud Infrastructure, and Automation
Anomaly detection over network telemetry wired to Cisco ISE and Symantec DLP, with a remediation engine that requires both high confidence and high severity before acting without a human. Appliance simulators included, so the whole stack runs without vendor licences.
Three studies in network operations: an LSTM autoencoder for unsupervised anomaly detection, a rule-based alert correlation engine that collapses an alert storm into one incident, and a Gymnasium environment for reinforcement-learning SD-WAN path selection. 73 tests, synthetic data only.
Twenty independent FastAPI services behind one compose stack with Prometheus metrics and a React dashboard. One is fully implemented — a catalogue normaliser that collapses five spellings of the same product into one key — and the other nineteen are the shared service template. The README says which is which.
Next.js and TypeScript application for AI-assisted security review of web and mobile projects, built on Firebase Genkit and Google AI. Personal product, developed January 2025 to January 2026.
Every open-source PostgreSQL hardening tool needs filesystem access, which rules out RDS, Aurora, Cloud SQL and Azure. dbaudit works from a connection string alone: it detects the platform, then gives remediation that is actually possible there. Emits SARIF to the GitHub Security tab. 114 tests, 97% coverage.
Production-ready Terraform modules for multi-cloud database provisioning across GCP and AWS. Includes auto-scaling, HA configuration, and GitOps-ready workflows.
Available for SRE consulting, cloud architecture projects, and infrastructure optimization