Kubernetes-Native AutoML Platform

Open-source platform for building, training, and deploying machine learning models on Kubernetes. From data to production with FLAML-powered automation.

Get Started View on GitHub

Terminal

# Create a training job
kubectl apply -f - <<EOF
apiVersion: klearn.klearn.dev/v1alpha1
kind: KLearnJob
metadata:
  name: churn-prediction
spec:
  dataSource:
    uri: s3://data/customers.csv
  taskType: classification
  targetColumn: churned
  flamlConfig:
    timeBudget: 3600
EOF

Why choose KLearn

Built for platform engineers and data teams who want ML without the complexity.

Kubernetes Native

Built on Kubernetes from day one. Custom CRDs for training jobs, models, and deployments. Scale effortlessly with your cluster.

Automated ML

Powered by FLAML for automated model selection, hyperparameter tuning, and feature engineering. No ML expertise required.

Full Observability

Integrated with Prometheus, Grafana, and MLflow. Track experiments, monitor deployments, and debug with confidence.

100% Open Source

No vendor lock-in. Self-host everything on your infrastructure. Apache 2.0 licensed with an active community.

GitOps Ready

Declarative configurations for training jobs and deployments. Integrate with ArgoCD or Flux for continuous ML.

Natural Language Interface

Chat with your data using Ollama-powered LLM. Build models, analyze results, and deploy with simple conversations.

Common Use Cases

From classification to time series, KLearn handles the complexity so you can focus on results.

Churn Prediction

Upload customer data, train a model, deploy for real-time predictions. All through the UI or YAML.

Lead Scoring

Automatically identify high-value prospects with classification models trained on your CRM data.

Fraud Detection

Build anomaly detection models that scale with your transaction volume on Kubernetes.

Demand Forecasting

Time series forecasting with automated feature engineering and model selection.

Join the Community

Whether you're a platform engineer, data scientist, or ML enthusiast, help us build the future of AutoML on Kubernetes.

1.2k

GitHub Stars

50+

Contributors

200+

Forks

Join Discord Star on GitHub