Looking for a job

Arnab Chowdhury

@arnabchowdhury3

Message

Data Scientist at AngelOne building multilingual LLM systems that classify 15,000+ monthly customer reviews.

India

Message

At AngelOne, I designed and deployed a multilingual customer feedback platform that classifies more than 15,000 monthly reviews using Databricks and OpenAI APIs. It reduced manual tagging effort by about 60–70% and cut issue triage from 3–5 days to under 24 hours.

At AngelOne, I also built campaign recommendation and uplift models to improve user targeting. A LightGBM click-through prediction model outperformed the previous model by 80%, contributing to a 66% increase in first trades among new users and a 40% boost in new-user activation.

At Fountain9, I developed demand forecasting solutions for enterprise retail clients, supporting inventory planning across perishables, non-perishables, and beverages. I optimized the model evaluation framework, reducing its runtime by 85%, and built ETL pipelines using BigQuery and NetSuite REST APIs.

At Ganit, I developed customer loyalty prediction and personalized promotion recommendation models for retail, alongside dashboards and reporting workflows. I hold a Master of Technology in Computer Science from Indian Statistical Institute.

Experience

Work history, roles, and key accomplishments

Current

Data Scientist 2

Current

AngelOne

Apr 2026 - Present (5 months)

Designed and deployed an LLM-based text classification system using Databricks and OpenAI APIs to categorize 15,000+ monthly customer reviews across multiple vernacular languages, reducing manual tagging effort by 60-70% and triage time from 3-5 days to under 24 hours.

Databricks OpenAI API LLM Text Classification NLP Jira Production Deployment

Data Scientist 1

AngelOne

Dec 2023 - Mar 2026 (2 years 3 months)

Built and deployed a neural network model for personalized omnichannel campaign recommendations, an uplift model using CausalML for offer optimization, and a CTR prediction model using LightGBM that outperformed the previous model by 80%, leading to a 66% increase in first trades and 40% boost in new user activation.

Neural Networks CausalML Uplift Modeling LightGBM CTR Databricks Model Deployment Monitoring

Data Scientist

Fountain9

Jul 2022 - Dec 2023 (1 year 5 months)

Developed and optimized machine learning-based demand forecasting solutions for enterprise retail clients, incorporating promotional, seasonal, and event-driven features, and built scalable ETL pipelines from BigQuery and NetSuite REST APIs.

Machine Learning Demand Forecasting ETL BigQuery NetSuite Python Feature Engineering

Senior Data Analyst

Ganit

Aug 2021 - Jul 2022 (11 months)

Developed a customer loyalty prediction model using decision trees, a recommendation engine based on PageRank, and automated dashboards for retail KPIs, while collaborating on geospatial clustering for supply chain optimization.

Machine Learning Decision Trees PageRank Recommendation Engine Business Intelligence Geospatial analysis Python