Starlight
AI inference
Private, OpenAI-compatible LLM inference on infrastructure you control.
→
Agent microVMs
Hardware-isolated sandboxes for AI agents, ephemeral or long-lived.
→
Kubernetes microVMs
Complete clusters on lightweight microVMs, Kubernetes unchanged.
→
Virtual machines
Run Windows and Linux VMs unchanged. No per-core licensing.
→
Industries
By industry
Critical infrastructure
Utilities, transportation, healthcare, and industrial OT.
→
Defense
Tactical edge, garrison, and forward-deployed compute.
→
Enterprise
Modernize VMs, containers, and AI on infrastructure you own.
→
By use case
VMware replacement
Same VMs, predictable licensing, no re-architecting.
→
Edge computing
Two-node HA, peer-mesh, atomic updates at the edge.
→
Private AI
Inference on your hardware. Data and weights stay yours.
→
Defense
Air-gapped, DDIL, and tactical-grade workload patterns.
→
Resources
Whitepapers
Architecture, comparisons, and reference patterns.
→
One-pager
Single-page brief you can hand off to a colleague.
→
Blog
What we are building, why, and other engineering topics.
→
What is edge-native?
Edge-native infrastructure for disconnected operations.
→
About
Portal
Ask AI
Ask AI about Starlight
Hand off context to your favorite LLM, or grab the
/llms.txt
link.
Open in ChatGPT
Prefilled prompt that points at /llms.txt
Open in Claude
Prefilled prompt that points at /llms.txt
Copy llms.txt link
mainsailindustries.com/llms.txt
→
View llms.txt
Markdown summary used by LLMs
Let's talk
01
/
05
01
→
What's your name?
Press Enter ↵ to continue.
OK
↵
Press Enter
Bot check
