busysloths-mlox 0.2.0.post2

Accelerate your ML journey—deploy production-ready MLOps in minutes, not months.

Sovereign AI Infrastructure. Open by Design. Slothfully Simple.

A configuration-driven control plane for deploying production-grade MLOps on your own servers — without cloud lock-in.

---

What is MLOX?

Cloud MLOps costs thousands per month. Setup is painful. Vendor lock-in is a trap.

MLOX is a calm, reproducible way to run production-grade ML infrastructure on your own servers or hybrid cloud. You define your stack in YAML, MLOX handles the rest — deploying services, managing secrets, and wiring dependencies across backends. Three interfaces (Web UI, TUI, CLI) share one inspectable config-driven core, while frontend-specific UI handlers are registered separately.

It's for engineers who prefer thoughtful systems over chaos. Backed by open source. Powered by sloths.

State of the Union (Sept 2025) — a short slide overview of what MLOX is, what problem it solves, and where it's heading.

---

Current Status

MLOX is in active alpha development (v0.x). Core infrastructure, all three backends (Native, Docker, Kubernetes), and the major services are functional. The project has been accepted at CAIN 2026.

We welcome contributors, early adopters, and honest feedback. If you hit something broken, please open an issue or reach out at [e-mail].

---

What Can You Do with MLOX?

Area	What's included
Infrastructure	Add/remove/tag servers; choose Native, Docker, or Kubernetes runtime; spin up single- or multi-node clusters
Services	Deploy, update, and remove services; centralized secrets; dependency wiring between services
Code	busysloths-mlox PyPI package with client integrations, SDK helpers, and example snippets
Lifecycle Management	Migrate, upgrade, export, and decommission services (planned)

---

Services Catalog

Category	Services	Status
ML Platforms	MLflow 2.x, MLflow 3.x, Airflow 2.x, Airflow 3.x	✅ Stable
Model Serving	MLflow MLServer	✅ Stable
LLMs & Inference	LiteLLM	✅ Stable
Vector & Feature Stores	Milvus, Feast	✅ Stable
Data & Streaming	PostgreSQL, Redis, MinIO, Kafka	✅ Stable
Observability	InfluxDB, OpenTelemetry	✅ Stable
Secrets & Access	OpenBao, Tiny Secret Manager, Docker Registry	✅ Stable
Kubernetes Add-ons	K8s Dashboard, Headlamp, KubeApps	🔄 Experimental
Cloud Integrations	GCP (BigQuery, Cloud Storage, Sheets, Secret Manager)	🔄 Experimental
Source Control	GitHub repository import	✅ Stable

---

Architecture in 30 Seconds

CLI     TUI     Streamlit Web UI     Other UIs
  \      |             |                /
   \     |             |               /
    +----+-------------+--------------+
                    |
                    v
      `mlox/application/use_cases/*`
         shared session-based logic
                    |
                    v
              `MloxSession`
   project + encrypted secret manager + infrastructure
             /                               \
            v                                 v
 secret-manager backend                `Infrastructure`
 (InMemory/TinySM/OpenBao/GCP)      topology for one project
                                            |
                                            v
                           `Bundle` = compute/server + services[*]
                                      |
                                      v
                    execution via `mlox/executors.py` + `mlox/execution/*`

MloxSession is the runtime center: it always carries project metadata, an encrypted key-value secret manager, and the current Infrastructure. The important shared application layer is mlox/application/use_cases/*; CLI already routes through it, and TUI/Web/future UIs should do the same. Infrastructure models topology through bundles, where each bundle groups one compute/server with its services. Anything executed on a compute goes through the execution layer, while compute capabilities already exist (git, docker, kubernetes, ...) and service capabilities are an emerging architectural direction.

YAML stays focused on deployable configs and Python build classes. Frontend-specific setup/settings components live in the frontend packages (mlox/view/*, mlox/tui/*) and are resolved through mlox/ui/registry.py, which keeps UI code out of service/server config definitions and creates a future extension point for plugin-provided UI handlers.

For deeper reading:

• Architecture Guide (humans) — codebase walkthrough
• Architecture Guide (agents) — high-risk areas and invariants

---

Quickstart

# 1. Install Task (https://taskfile.dev/installation/)

# 2. Clone
git clone https://github.com/BusySloths/mlox.git && cd mlox

# 3. Set up environment (creates conda env 'mlox-dev' with Python 3.12.5)
task first:steps

# 4. Launch the Web UI
task ui:streamlit

# 5. Or try the CLI
task ui:cli CLI_ARGS="--help"

See Installation Guide for a fuller walkthrough including Docker and Kubernetes setup.

---

Project Structure

mlox/
├── mlox/
│   ├── application/    # facade + session-based use_cases
│   ├── cli/            # Typer CLI package (root app + command modules)
│   ├── services/       # 20+ deployable ML services (one directory each)
│   ├── servers/        # Native and Ubuntu/SSH backends
│   ├── tui/            # Textual terminal UI + TUI-specific UI handlers
│   ├── ui/             # frontend UI handler registry
│   ├── view/           # Streamlit web UI + Streamlit-specific UI handlers
│   ├── session.py      # Runtime state & persistence
│   ├── infra.py        # Service/server graph
│   ├── config.py       # YAML loading + plugin discovery + UI handler lookup
│   ├── execution/      # backend/system execution helpers
│   └── executors.py    # remote task executor layer used by services/servers
├── tests/
│   ├── unit/           # Fast tests, no external deps
│   └── integration/    # Multipass VM tests
├── examples/           # OTel, MLflow tracking, DAG templates
├── docs/               # Architecture, installation, contribution guides
└── website/            # Astro landing page

---

Contributing

Sloth-Friendly Setup

# 1. Install Task (https://taskfile.dev/installation/)
# 2. Clone the repo
git clone https://github.com/BusySloths/mlox.git && cd mlox
# 3. Set up the dev environment
task first:steps
# 4. Install dev dependencies
pip install -e .[dev]

Run Tests

task dev:lint                   # flake8
task tests:unit:run             # unit tests (fast, no external deps)
task tests:integration:run      # integration tests (requires Multipass VMs)

Ways to Contribute

• Bug reports
• Documentation improvements
• Feature requests
• New service implementations
• Examples and tutorials

See CONTRIBUTING.md for the full guide and docs/WORKFLOW_QUICK_REFERENCE.md for how we use GitHub Projects, Milestones, and Issues.

---

Documentation

Document	Description
Installation Guide	Setup from scratch
Architecture (humans)	Codebase walkthrough
Architecture (agents)	High-risk areas and invariants
Contributing Guide	How to contribute
Workflow Quick Reference	Labels, milestones, PRs
Plugin Guide	External service and server config plugins
API Docs	Generated Python API reference

---

Sponsors

MLOX is proudly funded by:

Supporters

---

License & Contact

MLOX is open-source, distributed under the MIT License. Contributions are welcome and subject to the same terms.

We are looking for people invested in the problem we're solving. Say hello at [e-mail] or start a conversation in GitHub Discussions.