Talent's Information
-
Location
Mumbai, India
-
Rate
$13.0 per Hour
-
Experience
5.5 Year
-
Languages Known
English,Hindi
Available for
About SHUBHAM
A performance driven ambitious software developer with an extraordinary blend of and technical knowledge. Ability to communicate and motivate team members to enhance strategic goals and bottom line objectives. Creative problem solving and troubleshooting skills complemented by meticulous attention to details that will result in the success of an organization by developing new applications or improving existing ones.
Tech Stack Expertise
-
Microsoft SQL Server
SQL
0 Years -
Python
Python,TensorFlow,Numpy,Keras,Pandas,Django
0 Years -
HTML
HTML
0 Years -
CSS
CSS
0 Years
Work Experience

Data Scientist
- January 2018 - December 2022 - 5 Year
- India
Projects

Customer Conversion Analysis(Tourism)
- March 2018 - January 2019 - 11 Months
-
Data Transformation from client OLAP to OLTP system using pyspark
Data wrangling and statistical analysis of transaction data and performing various ML process for customer segmentation, forecasting of sale and Inventory using time series dataset
Used Machine learning technique like random forest and xgboost for customer conversion analysis
Worked in text pre-processing using regex, CNF and siamese network classifier model .
Conducted research on customer conversion analysis and put it into practice.
According to our client's description of the impact, their consumer base has increased by about 15%.

Music Propensity Model
- February 2019 - November 2019 - 10 Months
-
Demonstration to stackholders regarding capabilities of AI/ML for revenue generations
Converting clients requirements into data science terminologies
Data extraction from datalake using sql queries
Data wrangling and statistical analysis of transaction data and performing various ML process for customer segmentation, forecasting of sale and Inventory using music related dataset
Features analysis which is playing key role in clients revenue generation
End to end representation of concepts and EDA results to clients and implementing their feedbacks
Predicted a base of highly probable music subscription customers who can take subscription for music services
Demonstration of Model Performance to clients and showing impact of ML model statistically
My concept offers a 20% cost savings and a count increase of about 7%.

Generic Invoice OCR AI-NLP Product
- March 2020 - January 2021 - 11 Months
-
"Generic Invoice OCR AI –NLP product" is one stop destination to crosscheck vendor invoices .It is having capacity to process 1 Million invoices per day and providing anomalies
RESPONSIBILITIES:
Identification of increasing process efficiency of current audit process by use of AI/ML for cost and time savings and revenue generations for auditors of MNC(Indian)
Thought leadership by providing demonstration of ideation of Generic Invoice OCR AI –NLP product"
Creating technical roadmap to achieve committed results
Data extraction datalake using pyspark from hadoop distributed file system
Data wrangling and statistical analysis of transaction data and performing various ML process for customer segmentation, forecasting of sale and Inventory using music related dataset
Creating Computer vision related solution to extract all pdf files and putting for NLP and further processing
End to end NLP pipeline creation and demonstration of ideas in front of stackholders
End to End AI pipeline using Deep Neural network architecture along with customized encoders and decoders as per requirement
Deployment the same model over windows virtual machine and running it through django service with providing inputs and outputs
Statistical calculations to show impact of model over cost and time saving to clients
My product / concept has reduced manpower(Auditors) from 500 to 50 on above work

Churn Rate Prediction
- March 2021 - January 2022 - 11 Months
-
Footfall propensity analysis on a monthly level campaign which increases campaigning sales rate by 10- 15 % and reduces the customer churn rate by 8- 10%
RESPONSIBILITIES:
Gave ideation and pitch of project to clients for putting their data to work by analyzing what happened and why with help of AI/ML based Intelligent Decisions.
Thought leadership by providing demonstration of ideation of Footfall propensity analysis "
Creating technical roadmap to achieve committed results
Data extraction from datalake using sql queries and pyspark
Data wrangling and statistical analysis of transaction data and performing various ML process for customer segmentation, forecasting of sale and Inventory using Retail dataset
Created Time series forcasting related problem statement using extracted data
Feature Engineering to create predictor variables from the historic data which would help me capture the trend and seasonality in the data.
Plotted proper charts and visualization to show daily,weekly and montly footfall trend of retail customers to my client
Provided “Direct Forecasting” on above created data and demonstrated to client about sales rate increments
Deployment the same model over AWS and running it through django service with providing inputs and outputs
Statistical calculations to show impact of model over increase in sales and churn reductions to clients

Recommendation System
- March 2021 - November 2021 - 9 Months
-
Recommendation engine to increase sales by 10- 15 % based on the customer most bought category targeting and new suggestion. An image based recommendation is add on for clothing sales
RESPONSIBILITIES:
Thought leadership by providing demonstration of ideation of Recommendation engine to increase customers product sales
Creating technical roadmap to achieve committed results
Data extraction from datalake using sql queries and pyspark
Data wrangling and statistical analysis of transaction data and performing various ML process for customer segmentation, forecasting of sale and Inventory using Retail dataset
Created recommender system using extracted data
Feature Engineering to create predictor variables from the historic data which would help us to get impactful features.
Plotted proper charts and visualization to show impactful variables to boost sales of my client
Based on created models Suggesting recommendations
Deployment the same model over client premises machine and running it through flask service with providing inputs and outputs
Statistical calculations to show impact of model over increase in sales and churn reductions to clients

Vendor E-payment Cheque Verification
- June 2022 - November 2022 - 6 Months
-
Automated "Vendor e-payment cheque verification" for about 70k vendors/banks/partners with the frequency to process 10K invoices per day for the Audit client.
RESPONSIBILITIES:
Thought leadership by providing demonstration of ideation of Vendor e-payment cheque verification product to audit client
Creating technical roadmap to achieve committed results
Data extraction from datalake using sql queries and pyspark
Data wrangling and statistical analysis of transaction data and performing various ML process for Audit requirement
Created and automated pipeline for electronic payment verification
Deployment the same model over client premises machine and running it through flask service with providing inputs and outputs
Statistical calculations to show impact of model over cost and time reductions to clients
Soft Skills
Industry Expertise
Education

Computer Engineering in B.E
Maharashtra University- June 2015 - June 2017