Data Scientist in Process Mining and Digital Process Twins at Shell, Chennai
Job Description
This data scientist job in Chennai is with Shell’s Commercial Data Science team, in the Digital Process Twin portfolio, applying process mining, optimisation and AI/ML to process event data. The requisition number is R205059 and the posting closes on 30 September 2026.
At a glance
- Position: Data Scientist (requisition R205059), Job Family Group: Research and Development
- Where: Shell, RMZ Millenia, Chennai, India
- Pay: Not stated in the posting
- Experience level: Experienced professionals; 6–8 years in data science
- Apply by: 30 September 2026, the posting end date given by Shell
The Commercial Data Science team works with Shell’s business units on data-driven solutions. Within it, the Digital Process Twin Centre of Excellence advances process mining and analytics, and collaborates with academic institutions, industry forums and technology vendors on applied research in process analytics. The posting start date is 17 September 2026.
What the data scientist will do
- Develop and implement machine learning and optimisation models on process event data to predict, classify and improve business-process performance
- Apply process mining and AI/ML techniques to discover process patterns, identify root causes and quantify improvement opportunities
- Build and maintain process data models, integrating and transforming data from enterprise systems such as SAP, Salesforce and ServiceNow into analysis-ready pipelines
- Work with data engineers and software developers to embed models into Shell’s process-mining platform for scalable, real-time analytics
- Develop algorithms that combine process-mining insight with AI/ML models for anomaly detection, forecasting and prescriptive optimisation
- Implement MLOps and DevOps workflows for model lifecycle management, version control, deployment and monitoring using GitHub and CI/CD pipelines
- Evaluate emerging tools, algorithms and open-source frameworks in process intelligence and digital twins
- Embed Responsible and Explainable AI principles in model development and deployment
Why this role matters
Process mining is one of the few areas of industrial analytics where the data is already there — every enterprise system logs what happened and when — and almost nobody uses it well. Event logs are messy in ways that defeat standard tabular machine learning: cases have different lengths, activities repeat, and the interesting signal is in the order. Building models on that data, and then getting them into a production platform with proper MLOps, is a rarer combination than it sounds, and it transfers to any large organisation running enterprise systems.
What Shell is looking for
- Six to eight years of experience in data science, including hands-on development of AI/ML or optimisation models, and at least two years working with process event or behavioural data
- A Bachelor’s or Master’s degree in computer science, data science, engineering, mathematics or a related quantitative field
- A track record of building and deploying analytical solutions from concept to production
Pay and location
The posting does not state a salary. The role is based at RMZ Millenia in Chennai, is a regular full-time position, and sits in Shell’s Research and Development job family group within the Finance business unit.
How to apply
Apply through the official Shell careers posting for requisition R205059. Shell gives a posting start date of 17 September 2026 and an end date of 30 September 2026, so this one closes quickly. Check the posting for work-authorisation details before applying.
See more artificial intelligence jobs, or browse all research jobs on ResearchJobs.in.
Hiring institution: Shell
Official advertisement: shell.wd3.myworkdayjobs.com
How to prepare for this application
- Learn the vocabulary of process mining properly before the interview: event log, case ID, activity, trace, conformance checking, and the difference between discovery and conformance
- Have one example where you worked with event or sequence data rather than a flat table, and can explain how you framed the prediction problem
- Be concrete about production: which models you deployed, how they were monitored, and what you did when one drifted
- Expect questions on SAP, Salesforce or ServiceNow data structures; you do not need to be an expert, but you should be able to say how you would extract a usable event log from one
- Prepare a short, honest answer on explainability — the posting names Responsible and Explainable AI as a requirement, and generic answers are obvious
About Shell
Shell is a global energy company with operations across oil and gas, chemicals, power and low-carbon energy. It runs technology and business centres in India, including in Bengaluru and Chennai, covering research, engineering and digital work for its global businesses.
