Sanjoy Sinha

About Sanjoy Sinha

With over two decades of extensive experience, including 15+ years at IBM India, I am a seasoned Data Scientist and Automation Consultant specializing in Natural Language Processing (NLP), AI model development, and Robotics Process Automation (RPA). My expertise encompasses developing advanced AI-driven applications using cutting-edge technologies like BERT, GPT, and LangChain. I have a proven track record in enhancing user interactions through Watson Assistant chatbots across key industries such as telecom, FMCG, and insurance.My professional journey is characterized by strong command over machine learning, deep learning, and vector databases like FAISS and AstralDB, coupled with robust data analytics skills using Python and Tableau. I excel in end-to-end project management, leading cross-functional teams, and engaging clients to deliver solutions that drive business efficiency and informed decision-making.Significant projects under my leadership include automating multi-step queries for Vodafone UK in several African markets, creating predictive models for customer retention, and developing a high-accuracy fraud detection model. I have also managed large teams in high-stakes environments, as seen in my role leading RPA initiatives for British Petroleum UK, where we significantly enhanced operational efficiencies using Blue Prism.I am passionate about pushing the boundaries of AI and automation. I continuously seek to integrate innovative NLP solutions and advanced data analytics to revolutionize business processes and improve economic outcomes. I aim to leverage my industry knowledge and technical expertise to contribute to your organization\'s success.Let\'s connect and explore how we can drive innovation and excellence in your business operations with AI and automation.#DataScience #AIModelDevelopment #RPASpecialist #NLPExpert #AutomationConsulting

Experience

1. Data Scientist & Automation Consultant

   IBM

   Dec 2016 — Present · Kolkata, IN

   Project: Vodafone UK • Automated multi-step queries for countries like South Africa, Ghana, and Mozambique, providing a seamless user experience. • Built a BERT-based tool achieving 88% accuracy on customer feedback; improved performance with data preprocessing and hyperparameter tuning.• Developed an AI application using LangChain for enhanced language processing and RAG for accurate, knowledge-based query responses.• Designed an interactive front-end interface for real-time query input and accurate responses, improving user engagement.• Analysed state GDPs by industry, agriculture, and education, offering insights and recommendations for economic improvement.• Created a predictive model with 84% accuracy to identify churn risks, enabling targeted retention strategies.

Education

• Jalpaiguri Government Engineering College

  Bachelor of Engineering (B.E.), Civil

  1989 — 1993