Managed AI
made simple

A platform to create, manage, monitor and deploy
machine learning models.

screencapture-localhost-4200-dashboard-workspaces-f8b6e93e-2166-48b8-b983-47da81a4c1eb-use-cases-0152258c-e0f0-4a34-8729-abe75ce20408-pipelines-2020-07-30-17_35_57.png

4 Users - 1 Solution

navio allows for the collaboration of the entire AI landscape of a company. It helps to automate the process of creating, managing, monitoring and deploying machine learning models.

Business User

Leverage automatic machine learning
and get understandable results

Data Scientist

Upload & configure a model and deploy it
with a single click

Developer

Integrate a trained model into an
application via REST API

DevOps

Monitor the usage and
performance of a deployed model

How it works

#1

Choose your data set

Choose/upload a dataset or connect your database.

#2

Create your model

Use automatic machine learning or upload your custom model (MLflow).

#3

Understand your model

Investigate your model and get explanations of model results.

#4

Deploy your model

Put your model into production with a single click.

#5

Use your model

Integrate your model via a REST API and monitor its usage and performance.

ezgif-2-f879f2f97745.gif

Automated Machine Learning

You don't need to know machine learning to take advantage of it. Just upload your dataset and let navio do the work.

Determine the Goal

Decide if you are looking for a rapidly-trained model or a more robust model.

Automatic Transformations

Let navio do the transformations of the data to achieve better results.

Expert Mode

Advanced configuration for experts to fine tune settings and optimize results.

One-Click Deployment

Deploying machine learning-models into production can be tedious and complicated. Let navio handle this and deploy your models instantly via REST.

Easy Deployment

Deploy your model with one click. Define different models for different stages such as 'development', 'staging' and 'production'.

REST Interface

Interact with your model via your app or on your dashboard via REST.

Access Tokens

Handle access to your model via multiple access tokens and control quotas and requests.

ezgif-2-591b677623b3.gif
screencapture-localhost-4200-dashboard-workspaces-f8b6e93e-2166-48b8-b983-47da81a4c1eb-use-cases-0152258c-e0f0-4a34-8729-abe75ce20408-deployments-2020-07-30-17_26_39.png

Monitoring & Maintenance

Keep track of your model and the performance. See how many requests were sent and validate if the model is still accurate.

Model Accuracy and Performance

Monitor the performance and accuracy of deployed models.

Re-Training

Use new or additional data to retrain models and compare them with the original ones.

Monitoring & Error Logs

Detailed information of how frequently deployed models have been used, including results and error logs.

Support for Custom Models

You created your own custom model? Perfect! Just upload the model, deploy it and use it!

Support for MLflow

Upload your MLflow-models and deploy them for use.

Baseline vs Custom Model

Use navio as your baseline model, work on your custom model and compare the progress.

ezgif-2-00344dd97a3e.gif

What our clients say

navio took away my fear of contact with the entire Machine Learning domain.
Business User
navio supports us statisticians and mathematicians in bringing ML models to production.
Statistician

Ready to get started?

Contact us

Who is behind navio?

craftworks is a leading AI provider operating in German-speaking areas and has many years of experience in developing machine learning solutions for leading industrial companies. After years of work and research in this field it became clear that a platform like navio is missing on the market.

FAQ

Frequenty Asked Questions

Do I need data-scientists to use navio?

No, the software product is designed to be used by everyone. Automatic machine learning makes it very easy to use and understand. If this wasn't enough, our data-science team stands ready to support you in gaining the best results possible!

Can it run on my infrastructure?

Yes, navio can be run on-premise. Please get in touch with us for more information.

Are there any restrictions on the data types?

The automated machine learning supports numerical and categorical data in a tabular and time-series context right now. For hosting your models there are no limitations.

Do I need to install anything?

No, the software runs in the cloud. However, if you have specific needs or special data-security issues on-premise solutions are possible!