Experience

Boulder Hill Capital Management LP

• United States
• Financial Services
• 1 - 100 Employee

• Quantitative Developer

  • May 2021 - Present

  New York, New York, United States



Sauma Capital LLC

• United States
• Financial Services
• 1 - 100 Employee

• Data Scientist

  • Feb 2021 - Apr 2021

  New York, United States

• Data Science Intern

  • Sep 2020 - Dec 2020

  New York, United States

• Data Science Intern

  • May 2020 - Aug 2020

  New York, United States



Morgan Stanley

• United States
• Financial Services
• 700 & Above Employee

• Student Data Scientist

  • Jan 2020 - May 2020

  New York, United States • Worked here as part of Columbia University's IEOR E4524 - Analytics in Practice course • Examined and verified inflation and growth-rate based macroeconomic regimes based on equities and bonds monthly returns data (Dec. 1926 to Jan. 2020) through Exploratory Data Analysis, ANOVA tests, and t-tests • Automated search of macroeconomic regimes by introducing advanced analytics (decision tree) approach



McKinsey & Company

• Business Consulting and Services
• 700 & Above Employee

• Advisor, Data Science

  • Aug 2018 - Jul 2019

  Gurgaon, India • Improved server response time by 80% and reduced costs by 50% by restructuring design of ETL pipeline in AWS and presented insights to project sponsors • Implemented ETL pipelines in AWS Glue validating raw data in PySpark and SparkSQL with PostgresSQL • Led data team in deploying machine learning model in AWS employing Glue for ETL(Pyspark), s3 for saving model, and APIs for exposing model to end-user • Optimized revenue for oil refinery client by implementing a linear program in… Show more • Improved server response time by 80% and reduced costs by 50% by restructuring design of ETL pipeline in AWS and presented insights to project sponsors • Implemented ETL pipelines in AWS Glue validating raw data in PySpark and SparkSQL with PostgresSQL • Led data team in deploying machine learning model in AWS employing Glue for ETL(Pyspark), s3 for saving model, and APIs for exposing model to end-user • Optimized revenue for oil refinery client by implementing a linear program in Python (pulp) • Automated Simulation of manufacturing processes in Numpy, Python, accessible via APIs written in Flask (python) • Implemented ETL workflows in Alteryx for generating descriptive dashboard and forecasting demand in Tableau for retail client • Formulated solution strategies for problems at hand by collaborating with internal clients e.g. automating revenue optimization in cloud Show less



Pronto IT Labs

• India
• IT Services and IT Consulting
• 1 - 100 Employee

• Big Data Consultant

  • Jul 2016 - Jul 2019

  skills : Python, Java, Ipython, webpy Databases : Mongodb, Apache Cassandra Distributed Processing : Apache Spark, Apache Kafka Data Format : Apache Avro - Modeled a Naive Bayes classifier for news feeds categorization using Term Frequency Inverse Document Frequency (TF-IDF) score with 88% accuracy - implemented a web app for personalized and non-personalized recommendations based on users ratings using Alternate Least Squares (ALS) Algorithm - Implemented a data… Show more skills : Python, Java, Ipython, webpy Databases : Mongodb, Apache Cassandra Distributed Processing : Apache Spark, Apache Kafka Data Format : Apache Avro - Modeled a Naive Bayes classifier for news feeds categorization using Term Frequency Inverse Document Frequency (TF-IDF) score with 88% accuracy - implemented a web app for personalized and non-personalized recommendations based on users ratings using Alternate Least Squares (ALS) Algorithm - Implemented a data stream to capture clicks, conversion and server Logs using Spark Streaming And Kafka integration. Show less

• Python Developer

  • Jul 2014 - Oct 2015

  New Delhi Area, India Skills : Python, web.py Databases : MongoDB, Elasticsearch, neo4j, Redis - programmed CRUD operations for users and posts - Wrote global exception handler for webpy in python - implemented fuzzy search using Elasticsearch - implemented user-user and user-post linked structure using neo4j