Labelling data is essential to develop Machine Learning workflow. Another application is to enable automatic algorithm validation. In Schlumberger, we want to develop the testing of our new software through some automated validation workflow using labelled data.
While drilling an oil and gas well, abnormal events can occur that can jeopardize the safety of the people on the rig and the integrity of the rig. It is of primary importance to detect those events quickly to be able to mitigate them.
We are developing some new workflows to identify those events automatically, without any expert interpretation.
We need to quantify the detection performances of those workflows in terms of reactivity and reliability. We also need to have this quantification done at large scale to ensure our performance evaluation is robust and trustworthy.
We must have some experts’ labels references on a large number of datasets to be compared to the computed labels from the detection workflow.
Labelling accurately events at large scale is already a challenge. Indeed, it requires to be able to visualize time series in an easy and interactive way for the user. To be a success, this expert labelling campaign has to be done to facilitate the work from experts and ensure user adherence.
Your primary goal will be to create and design a webapp for the global Schlumberger initiative aiming at labelling abnormal events based on fit for purpose time series visualization.
Your secondary goal will be to test this webapp and ensure its deployment will be a true success inside the company. Thus, you will need to work in close collaboration with various experts all around Schlumberger (Clamart, Beijing, Houston, Cambridge…)
Targeted skills, and profile:
Final year engineering school or Master 2 level
Schlumberger is an equal employment opportunity employer. Qualified applicants are considered without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, or other characteristics protected by law.