Upamanyu

Upamanyu

Data Engineer/Data Scientist/Data Analyst
India

About Me

5+Years' Experience in Data Engineer/Data Analyst/Data Scientist with an overall IT experience of more than 10 years. ProficientinvariousMachineLearningAlgorithmsandStatistics- Regression, Simulation, Scenario analysis, …

Experience

Data Scientist

Skechers Inc
Jan 2021 - Present · 5 years 6 months

Developing ADF pipelines , Using Databricks and Developing ML models, Optimised ML models and deployed them in cloud. Used BI tools sich as PowerBI, Tableau and BO. Created Chatbhots. Experience in Data Engineering.

Machine Learning Engineer

Celandyne Software Solutions ,LLC
Jan 2021 - Present · 5 years 6 months

Working in a cross-functional team of engineers, product managers, and operation teams in Agile(Scrum) environment.
Created data pipeline in Azure Data Factory and Databricks and used Data Mining to clean datasets.
Utilized PySpark in Databricks to extract and load data and perform SQL queries.
Built machine learning models (training, validation, and testing ) with optimal solutions for data reduction, sampling, feature selection, and feature engineering
Created different Models-Trade Area Models, Customer Life time Analysis, Market Basket Analysis, Store Assortment, Price Assortment, Customer Churn, Recommendation system for 20000+customers daily.
For running Model Algorithms–classification/regression/clustering used Random Forest/K-means.
Constantly Evaluated the performance of the models, and made appropriate adjustments.
Developed Pricing strategy based on demand and elasticity of products.
Created data pipelines, for image data collection, annotation and used Computer Vision(Yolo ) for object detection.
Developed Business Intelligence solutions (Dashboards &Reports ) for different KPIs for the data coming from various custom systems and databases.
Engaged with product managers, analysts, and senior management regarding data pipelines, this improved productivity KPIs by 35%.
Used Anylogistix tool to optimize supply chain system ,Inventory Planning and Demand Forecasting.
Reduced Inventory cost and Transportation cost by US$ 20 million.
Automation reduced the manual work by 25%.

Data Scientist

Celandyne Software Solutions ,LLC
Jan 2020 - Nov 2020 · 10 months

Collaborated with Business(Finance)Subject Matter Expert(SME) to create enterprise level solutions
Migrated data from legacy systems(datamart) to DataLake by Sqoop in Cloudera and queried using HiveQL (Hive)
Developed Customer Delinquency Model by Calculating of PD, LGD, EAD for loans using credit risk models.
Converted data into metrics and reports and used Tableau for different charts to display small /subcategory datasets.
Delinquency model predicted customers who would default with 92% Accuracy.

DataScientist

Celandyne Software Solutions ,LLC
Sep 2018 - Dec 2019 · 1 year 3 months

Worked in a highly dynamic AGILE environment and participated in different Scrum meetings.
Used API to collect Data from Salesforce Sales Cloud.
Developed dashboards for reporting using Tableau to allow clients to track key metrics from their Saleforce.com databases to aid decision-making.
Used Rasa NLU Chatbot to create Sales Chatbot.
Did web scraping for Lead Generation.
Used Machine learning models to predict likelihood of payment.
Capable of running multiple scenarios and pickup the optimal solution.
Created a webframe work to run the python script in Flask.
Hyperparameter tuning and solved Imbalanced Dataset problem.
Created Models, compute clusters, Environment and Pipeline in Azure cloud.
Design and Implement Machine Learning flows from scratch and deploy the min to production.
Generate and maintain model metrics, model documentation
It reduced manual work of 20+hours per week of sales executive.
Better capture of potential customers.
Reduced Late payment by 60%.

Data Engineer

Celandyne Software Solutions, LLC
Jan 2017 - Jul 2018 · 1 year 6 months

Created different pipelines using different AWS components to ingest multiple TB of data from multiple disparate systems.
Capturedreal-timestreamofdataandingestedintoanalyticaldatasystems.
Used windowing functions for date string formatting conversions.
Implemented best practice processes and created cloud applications, and databases that have improved data accuracy for decision-making by 35%
Used AWS EMR to create environment to run various ML Algorithms.
Created pipeline that involved AWS SNS to notify stakeholders by SMS
Created tables in Amazon Redshift as DW and Dashboard in Amazon Quicksight for the systems and operations department.
Drafted functional and technical specification documents stipulating thepurposerequirements.
Removed manual labor, Optimized Fodder consumption and increased milk production.

IT Project Manager

Apollo Munich Health Insurance.
Apr 2016 - Jul 2016 · 3 months

Evaluated and optimized the project costs by working with various software consulting firms.
Analyzed the risks in projects and tried to mitigate them.
The 85% of digital products were released in the scheduled planned dates.

Data Engineer

Capgemini Technology Services India Limited
Jul 2014 - Apr 2015 · 9 months

Created Snowpipe to extract data from GCP.
Used databases using Snowflake and Reporting Services
Did Bulk and continuous Loading of data to different stages.
Loaded both structured and unstructured data into DW.
Added indexes, primary keys, did table partition for Optimization.
Developed stored procedures and SQL functions for ETL processing including staging and target processing.
Used Tableau as reporting tool.
Used visualizations in Tableau-such as dual axis, boxplots, treemaps, bubble charts, and waterfall charts-to solve complex problems and created interactive dashboards.
Used SQL queries in Tableau Desktop to validate.
Success of the project led to more projects from other regions of SabMiller Beer.

Data Analyst

Capgemini Technology Services India Limited
May 2013 - Jun 2014 · 1 year 1 month

Worked closely with business head of Nokia Siemens to create reports and Dashboards for BSS.
Successfully delivered multiple visualization using Bars, Lines, Gantt
Created effective DAX formulas and expressions, Calculated columns, Calculated Tables
Utilized Power Query in PowerBI to Pivot and Un-pivot the data model for data cleansing
Designed a PowerBI data model with multiple fact tables and dimensions depending on the business requirements.
Experience in publishing and scheduling refresh of PowerBI reports.
Assisted Product Managers designed and shared 50+ reports found common errors after performing extensive quality assurance testing.
The reports were designed to monitor Revenue collection.
It removed Revenue leakage by 90%.

BI Developer

Ericsson India Global Services Private Limited
Oct 2012 - Apr 2013 · 6 months

Successfully delivered multiple visualization using Bars, Lines, Gantt
Converted existing BO reports to tableau dashboards.
Created different KPI using calculated key figures and parameters.
Developed Tableau reports using Crosstabs, Heatmaps, Box and Whisker charts, Scatterplots, Geographic Map, Pie Charts , Bar Charts .
Developed donut charts and implemented complex features in charts like creating bar charts in tooltip.
Responsible for interacting with different stakeholders to identify requirements for reports.
The project helped clients monitor KPI’s for Performance Management of different towers.
It led to an increase in revenue of 12% for Telcel.

BI Developer

Ericsson India Global Services Private Limited
Jan 2012 - Sep 2012 · 8 months

Requirement gathering from Business Analyst and created the Universe using Universe Designer–IDT
Utilized Merged Dimensions feature to allow linking of two or more data provider (queries)
Created Contexts for resolving Loops and checked the cardinality for Integrity.
Created graphical representation of reports such as Bar charts, Pie Charts in Web Intelligence reporting tool
Did unit and integration testing for report validation.
Scheduled reports utilizing CMC.
The dashboard and reports led to15% increase in revenue for Tigo.

Data Engineer

CMC LTD
May 2008 - Dec 2011 · 3 years 7 months

Performed data analysis and data profiling using complex SQL on various sources systems including Oracle.
Involved in Data Analysis, Data Validation, Data Cleansing, Data Verification and identifying data mismatch.
Designed data model, analyzed data for online transactional processing (OLTP) and Online Analytical Processing (OLAP)systems.
Wrote and executed customized SQL code for ad-hoc reporting duties and other tools for routine.
Developed stored procedures and complex packages extensively using PL/SQL and shell programs.
Customized reports using PROCREPORT,PROC TABULATE and PROC
Migrated 20 billion Property and casualty policies in the whole process.

Skills

Big Data Data Evaluation Data Analysis Professional Data Warehouse Machine Learning Algorithms Python Pandas NumPy scikit-learn PyTorch spaCy Gensim PySpark Sqoop Hive Kinesis Databricks Flask Tkinter Power BI Tableau SAP BusinessObjects Clustering Classification Regression Object Detection Computer Vision Text Analytics Chatbot Web Scraping Operations Research AWS GCP Azure PyCharm AnyLogic Salesforce Oracle MS Access MySQL PostgreSQL Snowflake NoSQL Machine Learning Deep Learning NLP TensorFlow SQL Azure Data Factory Azure Machine Learning Azure ML SDK Hadoop Flume Spark MLlib Spark SQL
Report this Profile?