Monitaur is an AI governance software company, founded in 2019 to serve highly-regulated enterprises like flagship customer Progressive Insurance. The company delivers Monitaur Machine Learning Assurance Platform to help enterprises and their partners build and deploy responsible AI and machine learning models that business leaders, regulators, and consumers can trust.
The company’s integrated products – RecordML, AuditML, Monitor ML, and GovernML – empower AI/ML teams to deliver transparency, fairness, safety, and compliance for high-impact systems.
Monitaur is the first and only software built specifically to drive assurance and governance of machine learning, Monitaur’s suite of products enables stakeholders across the enterprise to deliver trust and confidence throughout the full lifecycle of their ML and AI systems.
Amazon VPC, Amazon ECS Fargate, Amazon S3, Amazon CloudWatch, Amazon API Gateway, Amazon Cloudfront, Amazon Cognito, Application Load Balancers, Terraform, Docker, Django, VueJS, Github, Github Actions.
Monitaur came to DinoCloud with the challenge of migrating the products’ environment from an Amazon ECS environment to a fully serverless ecosystem.
Monitaur’s top priorities were to ensure service and high availability so that platform disruptions did not occur and customers had a better experience. There were also pursuing a whole automated process for the deployment and management of their environments.
In the first place, we decided to assess the whole environment and deployment of the platform, identifying risks and pain points of improvement. During the review, we involved a lot of knowledge in the serverless AWS stack and the automation tools for infrastructure and CI/CD for the new architecture.
Once the main epics were identified, we set priority remediation objectives for the business, which were:
Once the project started, we decided to work together with the Monitaur team, using the Kanban methodology through daily and synergistic communication between the DinoCloud team of architects and engineers and the Monitaur developers and architects.
We decided to apply all the changes to the non-production environments until all the relevant tests were executed. Once completed, we were able to promote the new architecture in production traffic.
Monitaur understood that transparency, compliance, and safety were key factors in the brand and the image that the company wanted to transmit to its customers. With that baseline, their technical strategy needed to have a strong focus on security and reliability.
From its beginning, Monitaur decided to prepare a microservices-oriented architecture with common contracts for many of the products it were offering. Based on this, the team chose Django as their main back-end technology and VueJS for front-end development. Docker was the key technology for development and Amazon ECS for orchestration.
Over the years, complexity increased in the management of the routing and management of the traffic between the front-end and the back-end. Django UI was used until Monitaur decided to separate the layers. The team was convinced that serverless was the next step.
We worked with them refactoring the entire architecture to have a fully dis-aggregated and loosely decoupled puzzle, where each of the pieces where automatically deployed and environments were as much ephemeral as possible.
Each resource deployed in the AWS ecosystem was created using the good practices of IaC and Terraform, through a hierarchy of modules and environments to carry out this practice.
A well-optimized serverless architecture.
We achieve the automation of the creation, modification, and destruction of infrastructure using code, managing to apply DevOps methodologies in the management of said environments.
We migrated the JS serving from Amazon ECS to Amazon S3, adding security, performance, and reliability to the architecture.
We incorporated Amazon Cognito for the products’ authentication and authorization, among with Amazon API Gateway for the routing of the traffic.
Other tasks performed: