Case Study: Daupler optimizes MLOps and accelerates model delivery with ClearML

How ClearML Helps Daupler Optimize Their MLOps

Daupler, the maker of the Daupler RMS 311 response management system used by more than 200 cities and service organizations, needed a way to manage growing machine learning work with a lean team. Its data science group had to organize and label millions of data samples, maintain 15 and growing ML models, and support multiple inputs from calls, texts, geotags, photos, emails, web forms, social posts, and responder reports without wasting time on repetitive manual work.

Using ClearML’s MLOps platform, including ClearML Data, Pipelines, Orchestrate, AutoScaler, and Hyperparameter Optimization, Daupler automated dataset versioning, training, evaluation, deployment, and infrastructure scaling. ClearML helped the team quickly move data through the pipeline, reprioritize work when needed, and keep training resources only when required, enabling new concept models to reach MVP in about two weeks and refreshing production models in about a day.

Daupler

Heather Grebe

Senior Data Scientist