Ishman Preet Singh

Experience

• 

  Mu Sigma Inc.

  Jul 2016 - Dec 2020

  Lead and managed a 10-member offshore team to successfully cater to an assorted range of data science and data analytics for Europe’s largest sportswear and apparel manufacturer across:• CRM Analytics - Image Processing (CNN) – The business has been sending a lot of Dynamic and Static Newsletters to its consumers, and had been observing anirregularity in response for consumers. Vision was to maximize the email opens and conversions by identifying whatfeatures of the emails attracts consumers the most – The unstructured information in the form of digital content was converted into structured data by using ImageProcessing Convolution Neural Network based algorithms like YOLO + OpenCV for image feature extraction andTesseract OCR for text content extraction from the images – A decision tree based classifier was developed to define the best combination of features like orientation, text, context,font, etc. that has highest Click to Open Rate(CTOR) and A/B test was performed based on the recommended features«Technology»: • OpenCV • YOLO CNN • Tesseract OCR • Decision Trees• Financial Forecasting - Integrated Business Planning – The finance team wanted to set up a holistic top-line demand target setting process aligning all the business units forthe daily targets to achieve, taking returns and cancellations into consideration – Supported the finance team to find a standardized process by quickly building the detailed logic and churning outAlteryx Workflows to generate Demand to Net Sales targets for various markets– This helped the team in setting up a mechanism for business targets, aligned with all the business units to perform aretrospective of business performance«Technology»: • Alteryx • Exasol SQL • Python• Operational Analytics - Data Foundations– Provided technical leadership for enabling a single stop solution for all data needs for EU eCom by fabricating ready to use tables from existing data lake «Technology»: • SQL • Python • Shell Show less Actively supported 'Consumer Data Engineering' team of world's largets sports apparel & footware manufacturer and 'In-house Analytics Team' of world's largest thread manufacturing company through multiple analytical solutions• Smart Testing- Implemented a binary classification model to predict the wash-fastness and rub-fastness of thread lots, derived on various supporting factors like substrate material and thread thickness, thereby reducing the number of tests by 90%- Piloted the solution in China's plant and helped them achieve a labor reduction cost on fastness testing by 60%> Technology used: PySpark, Logistic Regression• Dynamic lot sizing- Assisted the clients by forecasting the demand of certain seasonal shade/substrate combinations, assessing the right time and lot size to be manufactured- This helped the client in optimizing the split between production, dormancy and inventory holding costs> Technology used: PySpark, Neural Networks• Consumer Data Engineering (Bot Identification)- Assisted in identifying and flagging bot users based on historical user activity and suspicious patterns of bot registration- This helped the client in excluding preferences of such users from any analysis and prevent targeting them for any promotional sales or exclusive offers > Technology used: Pattern Recognition, Python Show less Actively supported 'World Wide Services' team of the world's largest technology company and 'Risk and Finance IT' team of one of the bulge bracket investment banks through multiple analytical solutions• Services Insights BI- Worked on developing end to end automation of area specific sales reports providing technical account managers, an overview of the business health projections against the actual sales- Business health was made available to client as one-click sales reports on week over week basis bringing down the manual effort by 70%> Technologies used : SQL, Excel, VBA• Latin America BI- Assisted in building executive dashboards for customer support services team of Latin America division to assess the ongoing progress- Tasks involved working on massive data sets to generate meaningful insights and visualizations for sales excellence leads> Technologies used: SQL, Power BI• Services Account Growth- Performed driver analysis to identify high performing customers spread over a long span and understanding common behavior exhibited by them- Targeted various opportunities by creating a customer propensity model used to identifying low performing customers that had the potency to grow> Technologies used: Excel, R, SQL• Synthetic Data Generation- Worked on masking sensitive data on financial calculations by recasting it in a synthetic form, maintaining 100% business rule, code coverage and referential integrity- This version was circulated across geographies reducing the on-site development and testing cost by multi-folds> Technologies used: PL/SQL, R Show less

  • Apprentice Leader

    Jan 2020 - Dec 2020

  • Decision Scientist

    Mar 2018 - Dec 2019

  • Trainee Decision Scientist

    Jul 2016 - Feb 2018
• 

  Enquero

  Dec 2020 - Mar 2022

  • Client Engagement Manager

    Aug 2021 - Mar 2022

  • Lead Data Scientist

    Dec 2020 - Aug 2021
• 

  Adidas

  Apr 2022 - now

  Senior Data Scientist