Projects
Projects are labeled into Core, Individual, and Fundamental categories. Core projects show the step-by-step evolution of a single Infrastructure Platform—from a traditional search platform to a RAG platform and, ultimately, an AI agentic infrastructure platform. Individual projects cover separate studies and experiments that are not directly related to that platform. Fundamental covers foundational knowledge and concepts in infrastructure and other technologies.
The evolving AI infrastructure platform
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 later stage exploring tool-using agents, workload identity, short-lived credentials, service-to-service trust, observability, and controlled multi-agent execution.
Standalone projects and experiments
A structured way to turn technical topics into repeatable junior-level exercises: objectives, prerequisites, guided implementation, verification steps, troubleshooting, and review questions.
Small Kubernetes exercises covering manifests, Pods, Deployments, configuration, probes, troubleshooting, and the apply → inspect → debug workflow.
Foundational knowledge and infrastructure concepts that support multiple projects
A hands-on exploration of certificate chains, TLS handshakes, mTLS, Smallstep, SPIFFE/SPIRE, and how workload identity changes trust between services and agents.
Automated delivery built around safety, validation, and simple release control.
A reusable AWS foundation built around secure networking, environment boundaries, guardrails, and a scalable platform baseline.
Observability design focused on useful signals, clear dashboards, and practical alert quality.