Cloud engineering ⇔ AI infrastructure
CloudJi brings together my hands-on cloud and AI infrastructure projects, architecture decisions, and technical articles. It explores how systems are designed, secured, automated, monitored, and evolved—from traditional cloud platforms to AI infrastructure—while also covering other technologies and projects that interest me.
Latest from CloudJi
Begins with keyword search, adds embeddings and chunk-based vector retrieval, then introduces LLM generation with grounded citations. The frontend remains static while APIs and managed services provide dynamic behavior.
AWS VPC across multiple Availability Zones, CloudFront, API Gateway, S3 presigned uploads, SQS and DLQ processing, Python workers, OpenSearch BM25, IAM, monitoring, Terraform, and Docker.
A structured way to turn technical topics into repeatable junior-level exercises: objectives, prerequisites, guided implementation, verification steps, troubleshooting, and review questions.
A hands-on exploration of certificate chains, TLS handshakes, mTLS, Smallstep, SPIFFE/SPIRE, and how workload identity changes trust between services and agents.
Latest from CloudJi
Good architecture reduces the blast radius and improves recovery confidence long before an incident hits.
A practical framework for structuring a lab around one objective, clear prerequisites, verification, troubleshooting, and cleanup.
The value of observability is not collecting more data; it is creating clearer operational decisions.