Deploy your custom decision apps to Nextmv Cloud

Get your custom decision model into production in under 5 minutes. With decision apps on Nextmv Cloud, you can create a unique decision endpoint to run your model remotely and easily integrate it into your current infrastructure.

Decision applications on Nextmv Cloud are now generally available, allowing users to access their apps remotely using unique decision endpoints. Deploy custom decision models as a service and don’t worry about the infrastructure. Get started for free.

More optimization, less infrastructure management

“How do we get our model into production?” It’s a question that looms over many teams, from seasoned operations researchers to engineers entering the world of decision optimization. It’s a question that keeps projects from ever shipping at all. And when one does cross the finish line, a fear persists around updating the model in case it breaks during reimplementation and deployment. Many of us at Nextmv have certainly experienced the challenges that come with this process – and it’s one we don’t think you should have to worry about anymore. 

We created the Nextmv platform with a developer-friendly workflow designed to simplify the process of building, testing, and deploying decision optimization models. With a few commands, your custom decision model goes from local development to production infrastructure on Nextmv Cloud so you don’t have to think about translating models, standing up servers, or scaling resources.

Use the Nextmv CLI to package your custom decision model into an application that you can publish to Nextmv Cloud. 

Your custom model, your unique decision endpoint

Your custom model is what sets your operations apart. It represents your unique business rules and logic. Your teams live and breathe those requirements, refining the domain-expertise that makes optimization projects successful. Empower your decision scientists and devs by giving them tools to solve for any number of operational use cases –  without starting from scratch. Build faster by leveraging our templates, match business logic precisely by creating custom constraints and value functions, and then deploy with confidence. 

Run your custom decision application remotely by simply sending input data via an API call to your unique decision endpoint. 


Integrate decision optimization into your systems as easily as you would any other service. Connect your app directly to other components in your ecosystem (like databases, order management systems, and frontend applications) to power optimization under the hood for your internal operations as well as your external-facing services.

Don’t hesitate to update and innovate your models

We've all been there: you work really hard on a project and are ready for it to have a positive impact on the business. So you push it live and don't touch it (in case something breaks). But that's not how the universe works: conditions will change and models will need to evolve with them. 

Develop and iterate where you deploy. Run models using data structures and pipelines that mirror production so you can feel confident going from development to running in production. With Nextmv, there’s no language translation required across teams, so changes can be made quickly and transparently. 

The baton pass from decision engineer to infra engineer (and back again) just makes sense.

An outdated deployment process may hinder your team’s ability to ship improvements that have direct impacts on the business. Our dev-focused workflow allows you to make updates to your model locally and then deploy new versions to production with a few lines of code – incorporating continuous integration and continuous delivery into the optimization landscape. 

Validation, experimentation, and more to come

Coming soon, collaboration becomes even easier with OpenAPI pages. These autogenerated pages will document your app’s common REST API endpoints along with copy/paste-able code –  and will include on-page data validation for your schemas. See a preview in our deployment techtalk.

Next on the horizon is experimentation. We’ve built the Nextmv platform to not only make optimization models easy to get into production, but also easy to test in operational environments. See a preview in our benchmarking and testing techtalk.

Get started for free

Ready to see it in action? Spin up a free trial.

Have questions? Reach out to us or jump into the docs.

There’s more to come as we continue making custom model deployment a seamless experience. Stay in touch by signing up for our newsletter, registering for upcoming techtalks, or catching up our video content.

Video by:
No items found.