Data Analyst Python
Resume:
Jersey City, NJ-*****
GUNJAN SOOD
****.********@*****.***
EDUCATION
Ithaca, NY Cornell University Aug 2016 – May 2017
• Master’s in Applied Statistics & Data Science (GPA – 3.75/4)
• Graduate Coursework: DBMS & Data Mining, Big Data-Hadoop, Python Programming, Machine Learning Shimla, India Himachal Pradesh University July 2006 – July 2010
• Bachelor’s in Information Technology (GPA-3.9/4) SKILLS
• Languages and Tools: SQL, BigQuery, Python, DBT, Looker, Tableau, Google Analytics, JIRA, GitHub, MS Office
• Big Data Technologies: Amazon Redshift, AWS EC2, AWS S3, Google Cloud, ETL pipelines, Hadoop
• Machine Learning: Regression, Classification, Cross-Validation, Boosting, NLP EXPERIENCE
Data Analyst Paperless Post Dec 2017-Present
• Experienced in handling Product, Business, Marketing, and Customer Ops use-cases. Responsible for providing actionable recommendations and data driven insights to influence business decisions
• Analyzed customer patterns by creating conversion funnels and visualizations to understand user journeys. Provided recommendation to remove “login wall” which reduced user drop off rate by 30%
• Developed and automated KPI dashboard for C-Suite. These dashboards replaced the legacy excel reports, saved 4+ hours of manual work, and are now regularly used company-wide to monitor product pulse
• Took sole ownership of conducting and designing A/B test experiment for pricing tests. The experiments provided insightful recommendations on pricing which increased average customer value (ACV) by ~ +15%
• Created self-service reporting for Cust Ops to track open, delivered, bounce, and unsubscribe rate of emails
• Spearheaded end to end development of machine learning classifier (data collection, cleaning, feature engineering, model selection, training, testing, and deployment) to classify invitation cards into different categories, resulting in 94% model accuracy
• Designed data pipelines to extract and store NPS data from “Delighted” tool into our data warehouse
• Developed persona classifier to classify customers into different segments, helping management understand the user base and setting path for recommendation engine. The classifier increased F1-score by +10% Data Analyst
Moody’s Analytics
Jul 2017 – Nov 2017
• Created ETL in python to fetch data from API and store it in a database; reducing manual work by 30%
• Applied Natural Language Processing (NLP) using NLTK and topic modeling to infer topics of the documents Data Quality Analyst
Adobe Systems
Jul 2010 – Nov 2015
Adobe Premiere Elements (PRE)
• Implemented automation framework using Python to run 20 sanity test cases for unit testing and 180 test cases for full testing of the product to provide build status. Thus, reducing manual workload by 4 hours/day
• Involved in Agile sprint planning. Supervised QA work at the team level to ensure high quality deliverable at the end of the sprint. Also, responsible for communicating and providing QA reports to senior management
• Involved in regular code reviews with engineers and reviewing bug change lists before final code check-ins
• Boosted product sale by 10% (500K) by setting up Citrix environment (Linux) for European markets OPEN SOURCE - DATA CHALLENGES
• Yelp Dataset Challenge (Round 9): Classified users into 4 clusters: Elite, Slack Elite, Potential Elite, Non-Elite Implemented bag of words model (NLP) on 4.1M text reviews to find text patterns among these 4 clusters AWARDS AND EXTRA CURRICULAR
• Cornell Data Science (CDS) – Project team member of the largest Data Science club in Cornell
• Adobe Spot Award: A prestigious Adobe award offered to employees for going beyond work expectations
Contact this candidate