One-page case study
AI Platform Infrastructure — "Batcave" Personal ML Cloud
Solo-built private AI cloud on a Beelink mini PC — 56 containers, 36+ self-hosted services, 99.9% uptime. Runs local LLMs via Ollama, RAG pipelines, Authentik SSO, Tailscale VPN, and full Netdata monitoring.
Proof Points
56 containers
36+ services
99.9% uptime
60% lower hybrid-cloud latency
Challenges
- • Managing 56-container orchestration with zero-downtime deployments
- • Implementing zero-trust security across 36+ services
- • Optimizing hybrid-cloud caching reducing latency by 60%
- • Building automated security audit tooling with CWE/CVE scanning
Learnings
- • Production-grade infrastructure automation and MLOps
- • Multi-database architecture design and optimization
- • Advanced Docker networking and systemd service orchestration
- • Security engineering with SSO, VPN, and vulnerability assessment
Stack
DockerPostgreSQLMariaDBRedisPythonBashNginxAuthentikTailscaleNetdataLangChainOllama