Associate Data Science

You will be part of NPCI’sMarket Innovation team, working at the intersection ofadvanced machine learning, deep learning, graph AI, and Generative AIto build next-generation intelligent systems for India’s digital payments ecosystem.

This role focuses on solvingIndia-scale problemssuch asfraud detection, mule/AML risk modeling, transaction intelligence, and conversational AI, using bothclassical ML and cutting-edge AI architectures (LLMs, GNNs, Transformers, Agentic AI systems).

You will designend-to-end AI systems—from problem formulation, feature engineering, and model development toGPU-accelerated optimization and production deployment, ensuringlow latency, scalability, and robustness.

The role offers a unique opportunity to work on:

• Graph-based fraud detection systems
  
• Agentic AI & LLM-powered platforms (RAG, MCP, workflows)
  
• GPU/CUDA optimized AI pipelines
  
• Privacy-preserving and federated AI systems
  

You will collaborate withtop academic institutions (IITs/IISc)and cross-functional teams to push the boundaries of applied AI in financial systems.

• Job Title:Data Scientist – AI Engineer
  
• Division:NPCI Data Analytics – Market Innovation
  
• Education:B.Tech / M.Tech / MSc / MCA (PhD preferred) in CS, AI, DS, Mathematics or related field
  
• Employment Type:Full-time
  
• Location:Hyderabad
  
• Role Type:Permanent
  

• Develop and deployML/DL models(Logistic Regression, RF, XGBoost, NN, CNN, Transformers, GANs)
  
• Build models forfraud detection, AML, anomaly detection, transaction intelligence
  
• Work onimbalanced datasetsusing advanced sampling and cost-sensitive learning
  

• DesignGraph AI models: GNN, GCN, GAT, temporal graph networks
  
• Apply network analytics forfraud rings, mule detection, behavioral risk signals
  

• BuildLLM-powered applications(chatbots, complaint intelligence, document analysis)
  
• Implement:
  
  • RAG pipelines
    
  • Agentic workflows & MCP (Model Context Protocols)
    
  • Prompt engineering & LLM fine-tuning
    

• PerformEDA, feature engineering(temporal, behavioral, aggregated features)
  
• Work withstructured, semi-structured, and unstructured data
  

• Optimize models for:
  
  • Latency & throughput
    
  • GPU performance (CUDA-based optimization)
    
• Use libraries such as:
  
  • RAPIDS, cuDF, cuML, cuGraph, PyTorch Geometric
    

• Designcustom loss functions(weighted BCE, cost-sensitive)
  
• Applybusiness-aligned metrics:
  
  • Precision@K, Recall, ROC-AUC, PR-AUC
    
• Userobust validation techniques(cross-validation, time-based splits)
  

• Integrate models intobatch and real-time production systems
  
• Designscalable ML pipelines & APIs
  
• Monitor:
  
  • Model drift
    
  • Performance stability
    
  • Business impact
    

• Work withdata engineers, product teams, and business stakeholders
  
• Contribute toresearch, innovation, and academic collaborations
  
• Stay updated onlatest AI advancements (LLMs, Graph AI, Federated Learning)