Welcome!
Nodes orbit this profile summary. Open Skills for full graph controls, filters, and node search.
A sanitized view of my homelab operations: architecture, uptime signals, service categories, and telemetry patterns without exposing container names, ports, or internal endpoints.
Private compute and storage
Edge and backup operations
Live health snapshots
Private mesh and identity
Smart Hub
Continue exploring
Once you explore a few pages, this space becomes your quick return rail.
Featured build
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.
Latest note
A deep benchmarking session across two very different machines turned up an 8.5x speedup hiding in a single misconfigured setting, a genuine MLX advantage over GGUF, and one crash that took down an entire Mac Mini.
10 min read
New since last visit
AI Image Generator now runs on RTX GPU first, with automatic fallback
The image generator tries an on-demand RTX 3060Ti box first (faster renders, powered on only when needed) and falls back to the always-on Apple M4 Mac Mini if it's offline, with a health check cached for a minute so the switch never adds noticeable latency. A pre-flight VRAM check avoids overloading a shared GPU, and the UI now shows which hardware actually rendered each image.
Unique social preview cards for every page
Every route now generates its own Open Graph image from a shared branded template. Blog posts show their title, date, and tags; section pages show their name and summary. Links shared to LinkedIn, X, or Slack get a relevant preview instead of one generic card.
Jayakrishna Konda — ML/AI Engineer
I'm Jayakrishna Konda — an ML/AI Engineer and Data Scientist with 5 years building production ML systems, RAG pipelines, and self-hosted AI infrastructure. I design end-to-end solutions: from model training and LLM fine-tuning to cloud deployment and real-time monitoring. When I'm not shipping ML features at work, I'm operating Batcave — my 56-container private AI server running LLMs, RAG, and ML inference 24/7.
Production AI
5+ yrs
MLOps + DevOps
End-to-End
Batcave Uptime
99.9%
LLMs Self-Hosted
5+ Models