Manohar R

Sr Data Scientist @Cigna Healthcare

NY, US

Designed real-time AIOps anomaly detection models using unsupervised learning (DBSCAN, Isolation Forest, LSTM) to predict incidents and detect irregularities in application performance logs.• Built predictive models to estimate customer LTV, enabling better segmentation for upsell and retention campaigns.• Automated ETL workflows using Apache Spark and AWS Glue to integrate transaction, KYC, and customer interaction data across 12+ systems.• Deployed a customer support chatbot using Natural Language Processing (NLP) with GPT-based LLMs and prompt engineering for intent classification, achieving 90% accuracy in first-response resolution.• Implemented NLP pipelines for log analysis, alert deduplication, and intelligent event correlation, integrating outputs with Splunk and Datadog dashboards to enhance mean-time-to-detect (MTTD).• Enhanced NLP pipelines with language translation models (Transformer-based Seq2Seq) for multilingual client communication.• Collaborated with SRE and DevOps teams to automate root cause detection and incident triage across distributed microservices environments.• Exposed fraud detection and churn prediction models via FastAPI endpoints, enabling seamless integration with downstream systems and real-time scoring pipelines.• Constructed real-time dashboards in Power BI and Tableau to visualize fraud trends, transaction velocity, and model decisions.

Shanmugha Arts, Science, Technology and Research Academy

Bachelor of Technology - BTech, Computer Science

Illinois Institute of Technology

Master's degree, Data science