Raj Pravin

Data Scientist | Business Intelligence Engineer @ VML MAP

Experienced Data Scientist and Team Leader with a passion for solving real world problems. 7+ years of experience handling complex data in E-commerce, Marketing, and Finance Domains. Capable of using data-driven strategies to generate KPIs and insights for improving business performance.

Contact Me

About Me

My Introduction

I am an Indian national currently residing in the United Kingdom. I embarked on my professional journey as a Senior Analyst, specializing in the use of descriptive analysis to gauge product sentiment at Amazon India. My role involved transforming negative sentiments into positive ones by offering constructive feedback to vendors, drawing insights from data analysis.

Furthering my academic pursuits, I pursued a Master of Science degree in Data Science at the University of Westminster in the UK. During this time, I gained substantial experience in developing a plethora of machine learning models using freely available web data sources.

Subsequently, I joined Liberty Global as a Data Scientist, where my primary responsibilities encompass identifying and spearheading innovative data science projects. My role also involves guiding a talented team in the strategic implementation of data science projects, leveraging statistical methods to drive profitability for the company.

After two years at Liberty Global, I transitioned to a new and exciting opportunity at VML MAP as a Senior Data Insight Consultant. In this role, I continue to apply my expertise in data analytics and insights to help drive strategic decisions for the business.

15 Data Projects
Completed
Coming soon. Articles
Written

Skills

My Technical Level

Development

All About the Core

Python

80%

R

70%

SQL

90%

MS Excel

80%

Premeiere Pro

60%

Frameworks

Everyone Needs Support

NumPy

80%

pandas

90%

matplotlib

70%

scikit-learn

85%

NLTK

60%

seaborn

70%

Flask

40%

Machine Learning

Theory

Linear and Logistic Regression

95%

Decision Trees

95%

Ensemble Models

90%

Clustering

65%

Convolutional Neural Networks

80%

Recommendation Systems

75%

Natural Language Processing

65%

Exploratory Data Analysis

90%

Time Series

55%

Cloud and Engineering

Fly Fast & High!

AWS Redshift

65%

AWS S3

75%

GCP Storage

70%

GCP Big Query

70%

GCP Doc AI

70%

GCP Vertex AI

60%

Google Analytics GA4

80%

Google Tag Manager

80%

GCP Cloud functions

40%

Databases and Visualisations

Wow! Factor

MySQL

85%

AWS Redshift

75%

Tableau

70%

Power BI

80%

GCP Looker

80%

Qualification

My Personal Journey
Education
Work

Msc. Data Science

University of Westminster,London,UK
2021-2022

B.E. Aeronautical Engineering

Hindustan University, Chennai, India
2010-2014

Senior Data Insight Consultant

VML MAP
August 2024 - Present
What I did here

  • Extracted and organized data from Salesforce in Power BI using Power Query and DAX. Created clear and useful reports that met client needs and helped improve decisions, resulting in a €400K increase in campaign conversions.

  • Developed automated SQL scripts to transfer campaign data from Google Cloud Storage to BigQuery, ensuring seamless data ingestion and table creation. Scheduled the query to run every Friday, eliminating manual intervention and reducing effort by 12 hours per week, effectively saving headcount costs.

  • Conducted in-depth A/B testing analysis for email campaigns, optimizing customer segmentation and personalization strategies. This data-driven approach enhanced targeting accuracy, resulting in a measurable revenue uplift of 10% and increased customer engagement.

  • Created a script in dbt that reduced processing time and storage usage in BigQuery for the IKEA Family Dashboard project, lowering data size from 45TB to 32TB.

Data Scientist

Liberty Global
October 2022 - July 2024
What I did here

  • Utilized GCP tools (Doc AI, BigQuery, looker), Python, and SQL to automate invoice-PO matching, which help in reducing 10 positions in a team

  • Extracted Data from 2+ sources and converted them into one combined source for reporting purposes using ETL techniques and built Business intelligence dashboard in looker, saving 2 FTE of manual reporting work.

Business Intelligence Engineer

Amazon
August 2018 - September 2021
What I did here

  • Applied "Market Basket Analysis/Association rule mining" on a large customer data to discover the patterns for bundle packing which helped in increasing the sales to 17%.

  • Performed efficient feature engineering on a dataset with 148 columns using Python libraries, including NumPy, Pandas, and Seaborn. This optimization reduced model training time from 1 hour to 35 minutes.

  • Analyzed and reported quarterly revenue figures to internal stakeholders using Looker dashboards

  • Applied Content-based recommendation system algorithm for the books category which increased in the sale of 8%.

Data Analyst

Amazon
June 2016 - July 2018
What I did here

  • Analysed 30000+ responses to a consumer feedback to evaluate brand perception and given feed back to the vendor to improve the quality of the product which decreased the negative comments from 60% to 23%

  • On Boarded Vendor calling initiative , which improved the vendor TAT by 45% resulting in immediate resolution for existing issues on Catalog.

  • Finding and replacing the invalid data from the detail page via Python and this project helped in headcount savings of 3.17 FTE.

Portfolio

My Projects

Coupon Prediction

Using ML predict a Person who can claim the Coupon

  • Using the data given I have applied various ML models to predict the person who is eligible to claim the coupon and deployed it using flask.

  • Tech Stack


    View Code

    Shoe Price Prediction

    Predicting Shoe price using Machine learning

  • This is a demo project to elaborate how Machine Learn Models are deployed on production using Flask API

  • Tech Stack


    View Report

    CAR Data Dashboard

    Dashboard on CAR dataset Via Looker

  • Designed an analytical dashboard within Looker to comprehensively analyze factors influencing pricing variations in the specific car dataset, discerning the determinants behind both lower and higher price points for individual vehicles.

  • Tech Stack


    Research Papers Referred

    View Dashboard

    Market Basket Analysis

    University Project

  • Leveraged the Apriori algorithm to conduct a Market Basket Analysis, extracting valuable association rules from a specific dataset. This analysis provided strategic insights into effective product bundling and sales strategies for optimizing the website's offerings.

  • Tech Stack

    Research Papers Referred

    View Code

    Clustering - Unsupervised Learning

    Identify the main groups of countries and their global happiness characteristics using Clustering techniques - University Project

  • Used clustering techinques such as K-means and Agglomerative to find out the groups that are globally happy. Also, I ued Elbow method and Silhoutte score to find the right number of clusters

  • Tech Stack

    View Code

    Diabetes Prediction

    Predictive Modeling for Diabetes Detection Using Data Mining and Machine Learning

  • Utilizing the provided dataset, I employed a range of machine learning models to forecast the likelihood of individuals having diabetes. My project involved the exploration of diverse predictive techniques, contributing to the development of valuable insights for diabetes risk assessment.

  • Tech Stack

    View Code

    Tesla Stock price

    predicting the stock price via time-series analysis

  • I conducted a stock price prediction project specifically focused on Tesla's performance. To achieve this, I harnessed historical Tesla data and incorporated Long Short-Term Memory (LSTM) models, a cutting-edge deep learning technique. In my project bio, I can highlight that I successfully applied LSTM models to forecast Tesla's stock price, showcasing my expertise in financial data analysis and advanced predictive modeling techniques.

  • Tech Stack

    View Code

    Iphone data analysis

    Data analysis task using Python

  • Apple iPhones stand as prominent contenders in the global smartphone market, renowned for their widespread popularity. Despite the intense competition from various smartphone brands in India, where cutting-edge technology is available at a fraction of the iPhone's cost, the sales of iPhones continue to thrive. This article delves into the analysis of iPhone sales in India using Python, offering insights into the factors contributing to their success in this competitive market.

  • Tech Stack

    View Code

    A/B Testing

    Data analysis task using Python

  • A/B Testing means analyzing two marketing strategies to choose the best marketing strategy that can convert more traffic into sales (or more traffic into your desired goal) effectively and efficiently. A/B testing is one of the valuable concepts that every Data Science professional should know.

  • Tech Stack

    View Code

    Hunt And Seek

    Game Data analysis task using Python

  • The data provided contains information about a random selection of 10,000 players who installed one of our apps, Hunt and Seek (App store pages: Android and iOS), over a 14 day period and the objective is to find out ‘Organic’ players have better gameplay Key Performance Indicator (KPI) metrics than ‘Paid’ players

  • Tech Stack


    Research Papers Referred

    View Dashboard View Code

    Covid-19 Data Dashboard

    Quicksight Dashboard to show how covid occured in 2020 and 2021

  • The data provided contains information about covid which has spreaded among the varies countries.

  • Tech Stack

    View Dashboard

    Kody Pay Looker Dashboard

    Analyzing Payment Dynamics and Profitability.

  • Created a detailed report summarizing the total fees paid by a customer, Kody’s financials, payment distribution by card type, and hourly payment trends, including a formula to estimate future PSP fees based on card type. Provide additional insights derived from the customer’s payment data.

  • Tech Stack

    View Dashboard

    Just Play Data Analysis

    Analyzing Ad Spend data and Providing Insights.

  • In this project, the goal is to explore the given dataset, generate key metrics, and derive actionable insights. The analysis will focus on addressing data cleanliness, ensuring proper joins, and applying appropriate aggregation levels. The findings and the detailed process followed during the analysis will be documented and shared.

  • Tech Stack

    View Dashboard View Code

    Tonies Data Analysis

    Analyzing Playback Data.

  • Data Analysis and finding out which field is important in terms of customer retention

  • Tech Stack

    View Code

    Sentiment Analysis

    Natural Language Processing

  • Conducting a comprehensive sentiment analysis on the Amazon phone dataset, I utilized TF-IDF vectors to assess product sentiment and predict star ratings. Leveraging this analysis, I provided actionable feedback to vendors, aiming to mitigate negative sentiments and enhance overall customer satisfaction. In my project bio, I can emphasize my proficiency in sentiment analysis techniques applied to large-scale e-commerce data, showcasing my ability to derive valuable insights for improving product perception and customer experience.

  • Tech Stack

    View Code

    Certifications

    Extra Courses I have Undertaken

    Intermediate SQL for Data Scientists

    Expiry Date: Does not expire

    View Certificate

    Executive Data Science course from great Learning

    Expiry Date: Does not expire

    View Syllabus

    Developing Data Models with LookML Completed

    Expiry Date: Does not expire

    View Certificate

    Contact Me

    Get in Touch

    Call Me

    +447747265013

    Location

    London, UK