Case Study: SLB achieves faster, more accurate well data extraction with Snorkel AI

How SLB uses Snorkel Flow to enhance proactive well management

SLB, the world’s leading provider of technology and services for the energy industry, needed a faster way to extract critical geological and field data from highly varied, unstructured PDF reports. Its team struggled with manually labeling training data, poor collaboration between domain experts and data scientists, and document formats that made named entity recognition difficult at scale. SLB turned to Snorkel AI and Snorkel Flow to help automate information extraction for proactive well management.

Using Snorkel AI’s data-centric AI workflow and programmatic labeling in Snorkel Flow, SLB and Snorkel experts built an AI application that could identify 18 industry-specific entities across 15 document structures. In just three days, the team reached an 85% F1 score and improved it to 91.4% with rapid iteration, while cutting report processing time from 1–3 hours to just seconds. The solution also improved generalization by 47% and reduced document processing time by 99%.

SLB

Swaroop Kalasapur

Head of SLB Technology Innovation Center