You are struggling to prepare datasets for deep learning applications.
We automate data preprocessing, assist in feature engineering and scale on large datasets.
You are having difficulties training, evaluating and serving deep learning models.
We make model development a concise and coherent process.
The Core Engine is a consolidated interface to build tailored pipelines, from data ingestion, to training-evaluation, and finally serving. Quickly construct production-grade pipelines with our easy CLI, API or Web UI. Focus on building machine learning models with the tools you already know, and rest assured that your workflow is collaborative, scalable and production ready.
We understand that Deep Learning development involves repetitive experimentation. The Core Engine will monitor your work meticulously so that you never have to do the same thing twice. Avoid rewriting code that you have written before, track execution steps that have already happened, and skip redundant and expensive computations.
The Core Engine is developed API-first with ops and automation in mind. Integrate our solution into your infrastructure via API, CLI or through our web application. Pipeline runs can be automated with no hassle, and all relevant artifacts are versioned by default.
With big enough data, it can take hours to crunch through data in one single machine. The Core Engine uses distributed data processing technologies (Apache Beam) for efficient execution, reducing hours of computation to just minutes.
We built it because we scratched our own itch while developing for deep learning in production.
We plan to launch the CLI in March 2020. The Web UI will be ready soon after that.
Currently, we run the workloads on the Google Cloud Platform. We are working on having it integrated into our user’s cloud platforms.
We keep it safe and isolated. The storage and workloads run only in datacenters compliant to ISO 27001, ISO 27017, ISO 27018 and GDPR.
Both are developed by the same folks. The Asset Optimization Platform (AOP) is a product aimed at enabling industry use-cases like predictive maintenance and damage detection. The Core Engine used to be the internal platform that powered the AOP, until we decided to release it as its own product.
Stay tuned for our pricing plans as soon as we launch!