Senior Data Scientist

Caterpillar Caterpillar · Industrial · Bangalore, Karnataka

Senior Data Scientist at Caterpillar to develop and scale a Manufacturing & Supply Digital Platform using AI and NVIDIA technologies. The role involves architecting ML and synthetic data pipelines, developing generative and optimization models for manufacturing processes, and deploying them on digital twins for closed-loop optimization. Requires expertise in ML, deep learning, optimization algorithms, and Python, with a focus on industrial analytics and data-driven decision making.

What you'd actually do

  1. Architect end-to-end ML and synthetic data pipelines to drive continuous improvement in virtual manufacturing, using OpenUSD/Omniverse simulations.
  2. Develop advanced generative and optimization models for scheduling, material flow, quality prediction, and process control; deploy on digital twins for closed-loop optimization.
  3. Evaluate and implement modern optimization algorithms: stochastic search, evolutionary/genetic, reinforcement, and multi-objective optimization.
  4. Lead model validation using industrial analytics metrics (yield improvement, downtime reductions, forecast accuracy).
  5. Establish standards for data/model lifecycle management, production monitoring, and feedback from physical operations.

Skills

Required

  • machine learning
  • deep learning
  • composable model services
  • ensembles
  • retrieval augmented generation
  • multimodal orchestration
  • rigorous evaluation methods
  • classical optimization
  • stochastic search
  • evolutionary algorithms
  • genetic algorithms
  • reinforcement learning
  • multi-objective optimization
  • Python
  • high-performance analytical workloads
  • simulation
  • deployment
  • manufacturing process analytics
  • multivariate analysis
  • defect analytics

Nice to have

  • Team/project leadership
  • publications
  • patents in industrial ML or optimization
  • cloud-native deployments
  • industrial IoT/edge analytics
  • data pipeline automation
  • plant/line telemetry
  • quality analytics
  • maintenance analytics
  • logistics analytics
  • scheduling analytics
  • OpenUSD-powered asset graphs
  • real/virtual process data integration
  • digital twins

What the JD emphasized

  • Deep expertise in machine learning and deep learning techniques
  • Deep knowledge of manufacturing process analytics

Other signals

  • Develop advanced generative and optimization models for scheduling, material flow, quality prediction, and process control; deploy on digital twins for closed-loop optimization.
  • Architect end-to-end ML and synthetic data pipelines to drive continuous improvement in virtual manufacturing, using OpenUSD/Omniverse simulations.
  • Evaluate and implement modern optimization algorithms: stochastic search, evolutionary/genetic, reinforcement, and multi-objective optimization.