Client background
Our client provides an enterprise cloud platform built on a private network. It draws on nearly 20 years of work in security, network architecture, collaboration, AI, and open source software. Customers in more than 150 countries use it to modernize their computing environments.
Business challenge
The oil and gas market needed a platform that could process different types of input files, extract meaning from them, and run advanced analytics for specific geodesic problems. The client set out to build it, expand the services it offers, and increase usage.
They lacked specific resources, so they looked for a partner who could provide a core team that would grow with the business. They also needed well-documented solutions that allow burst capacity as new opportunities arise. SoftServe joined as a development partner, bringing machine learning and cloud compute experience, and as a go-to-market resource for major oil and gas opportunities.
Project description
SoftServe started with a machine learning (ML) workshop in Austin, Texas. The workshop captured domain knowledge and existing use cases and scoped the ML resources needed. The team then built the platform in a series of development sprints.
The use cases are primarily ML-based and cover above-surface and subsurface work, including predictive maintenance, seismic data interpretation, and offshore IoT data analysis.
The dedicated delivery team included two system architects, three to five ML engineers, one project manager, and one business analyst. They worked with GCP AutoML and CloudML, Kubeflow, Kubernetes, Vision API, NLP, deep neural networks, and generative adversarial networks. The platform handles both structured data (seismic data and well logs) and unstructured data (PDF and PPT files), which makes it easier to analyze, navigate, and search.
Value delivered
The platform combines advanced analytics, platform features, and cloud scalability. Key benefits include performance optimization, new features, reworked approaches using current techniques, and moving ideas into production.
It will automate much of the routine manual work geophysicists do for oil and gas businesses. It ingests, analyzes, and stores large volumes of data efficiently. The platform will speed up building basin models and other reports by 10-30x.
The client continues to work with SoftServe to add features, improve their quality, and shorten time to market for its customers.
