An operator providing a multi tenancy solution which allows DevOps teams to request a context for their project, which we like to call a 'Project as a service', e.a. Paas.
Kubernetes operator for self-hosted LLM inference across a heterogeneous GPU fleet: NVIDIA CUDA, AMD Vulkan, and Apple Silicon Metal. Runtimes: llama.cpp, vLLM, TGI, mlx-server. Multi-GPU sharding, model caching, OpenAI-compatible endpoints. Apache-2.0, run across homelab and on-prem fleets, actively developed.
Kubernetes operator for deploying and managing LiteLLM AI Gateway. Declarative CRDs for models, teams, users, and virtual keys with bidirectional config sync, SSO/SCIM user provisioning, and OLM support. Built with Operator SDK.
Native Tailscale egress for groups of Kubernetes pods — reaching tailnet peers, advertised subnet routes, app connectors, and exit nodes — without running tailscaled in every pod