Case Study: Travtus streamlines machine learning workflows with Devoteam and Amazon SageMaker

Travtus cuts model training time with Devoteam using SageMaker pipelines

Travtus, a London-based AI research and development company, faced challenges with the time-consuming management of training data, ML Ops, and production model monitoring for its AI teammate, Adam. To explore a more efficient solution, they partnered with Devoteam to assess the feasibility and benefits of migrating their workflow to Amazon SageMaker.

Devoteam implemented a proof-of-concept project using Amazon SageMaker to automate the end-to-end machine learning workflow. The solution included creating a PyTorch pipeline to fine-tune a pre-trained model and implementing CI/CD best practices. This provided Travtus with a scalable, cost-effective platform that streamlined their processes, improved model accuracy and reliability, and introduced robust lineage tracking and automated model retraining.

Travtus