Gowtham Senthil

Developing a Data Intelligence Platform for PHM (Prognostics and Health Management)

I'll be focusing on the development and validation of ML models for anomaly detection, predictive maintenance, and battery health estimation with a particular emphasis on Remaining Useful Life (RUL) prediction for BESS and battery products.

IoT Fingerprinting & SMS Analytics - Revene Assurance -> Developed a big data ML pipeline that uses PySpark and Hadoop, and deployed it on Cloudera clusters in a Linux environment, enabling real-time predictions for downstream applications and reducing processing time by 40% while maintaining 90% accuracy -> Performed A/B testing of anomaly detection model variants using Python to compare performance metrics, enabling the guided selection of the production model -> Developed a generative AI chatbot that democratizes data access by translating natural language into optimized SQL, leveraging a Neo4j knowledge graph and RAG to provide the LLM with relevant context -> Enhanced contextual reasoning of the chatbot with LangChain pipelines, achieving 99% retrieval accuracy and 30% lower token usage. -> Conducted and managed project presentations and product demonstrations for Fortune 500 clients such as T-Mobile and Etisalat, leading to positive client feedback and increased interest in our solutions

Developed ETL workflows and built analytical dashboards using Java, SQL, and Tableau, applying data structures and agile methods; the dashboards accelerated stakeholder reporting and earned the Team Star award in Q3 2023

-> Collaborated with economists to integrate financial data using Python and Pandas, and helped build India’s largest budget-document repository, processing over 18,000 files -> Applied fuzzy matching algorithms to facilitate data matching and mapping, improving data integration processes

Worked on Text-to-Speech Engines and contributed towards developing an Interview scheduling bot that automates the process of cold calling.