$$$
{{ $t($store.state.user.experience_value_in_dollars) }}
Senior
{{ $t($store.state.user.experience_search_name) }}
0
jobs
With more than 5 years of data analytics and data science experience leading multiple teams for data science projects and collaborating across teams I am capable of taking up your next data science project
Jomy Sebastian
,
Cherthala, India
Experience
Other titles
Skills
I'm offering
A Data Scientist with more than 5 years of industry experience in building Traditional BI, Machine learning and Deep learning projects. Experienced in leading multiple projects and working with teams across technology domains. A quick learner who enjoys learning new tools and technologies.
Markets
United Kingdom
Links for more
Once you have created a company account and a job, you can access the profiles links.
Industries
Language
English
Fluently
Ready for
Larger project
Ongoing relation / part-time
Full time contractor
Available
My experience
2019 - ?
job
Senior Data Scientist
Vonnue Innovations Pvt Ltd.
Movie Recommender - In house project for scraping movie data and build recommendation using Graph
algorithms in Neo4j. Build a serverless backend architecture by using Neo4j Rest API.
Google Ads Scraper - A module to scrape google ads and organic result. Scraped google ads by automatically
generating keywords and using AWS lambda invocations for scraping each Google Ad Copies and organic
results.
Metadata Tracker - A python module for tracking metadata of a clients service and programmatically generate
multiple excel status report weekly by comparing against its contend provider to keep it up to date.
Cloud Pipeline Migration - Build solution to cut cloud storage cost by converting json/csv to parquet format in AWS S3, which also lead to efficient querying against huge data storage. Storage size reduced by 75%.
Churn Prediction - Build churn prediction model using XGBoost against a content provider user base with 82%
accuracy. Designed end-to-end data flow from cloud to predicting churn and saving results back to cloud.
User Segmentation- Build user segments using GMM model. Designed and developed end-to-end data
pipeline. Manually done data profiling for each sku and region.
algorithms in Neo4j. Build a serverless backend architecture by using Neo4j Rest API.
Google Ads Scraper - A module to scrape google ads and organic result. Scraped google ads by automatically
generating keywords and using AWS lambda invocations for scraping each Google Ad Copies and organic
results.
Metadata Tracker - A python module for tracking metadata of a clients service and programmatically generate
multiple excel status report weekly by comparing against its contend provider to keep it up to date.
Cloud Pipeline Migration - Build solution to cut cloud storage cost by converting json/csv to parquet format in AWS S3, which also lead to efficient querying against huge data storage. Storage size reduced by 75%.
Churn Prediction - Build churn prediction model using XGBoost against a content provider user base with 82%
accuracy. Designed end-to-end data flow from cloud to predicting churn and saving results back to cloud.
User Segmentation- Build user segments using GMM model. Designed and developed end-to-end data
pipeline. Manually done data profiling for each sku and region.
Content, Social media ads, UP, Backend, Lambda, Google, Neo4j, Data Storage, Storage, Serverless, It, Architecture, Excel, Algorithms, Service, ADS, Cloud, REST, JSON, REST API, AWS, API, Backend, Python
2018 - 2019
job
Data Scientist
Techvantage Systems Pvt Ltd.
Smartphone Screen Crack detection - Build a screen crack detection model using CNN to detect if a smart
phone screen is cracked or not. Model had a high 85% accuracy against the test image set.
Cattle Insurance Project - Digitalized cattle insurance project. Devised complete work flow for detecting cow
muzzles, extracting features and comparing. Used MRCNN model to detect cow muzzle with 98% accuracy. Used
a CNN model to classify images with/without a cattle. Used 2d feature extractors like SIFT, SURF, BRIEF and ORB in python's OpenCV library to extract muzzle features and Flann matcher for pattern matching.
Face Detection in Live stream- Created a face detector and recognizer from video stream using CNN model and OpenCV in python.
Fraud detection and investigation - A fraud detection and investigation project PoC on transaction data using
multiple platforms like bank transfer and UPI. Delivered SQL database queries for tagging suspicious activities on transaction data. Single handedly developed schema and solution for investigating fraudulent activities using
Neo4j's graph database on millions of records.
Incentive Management System- A platform for analyzing incentives for sales executives. Build quota
recommendation system using Random Forrest regressor in python for an incentive management project.
Created descriptive analytics and dynamic dashboards for What-If analysis using Tableau dashboards.
phone screen is cracked or not. Model had a high 85% accuracy against the test image set.
Cattle Insurance Project - Digitalized cattle insurance project. Devised complete work flow for detecting cow
muzzles, extracting features and comparing. Used MRCNN model to detect cow muzzle with 98% accuracy. Used
a CNN model to classify images with/without a cattle. Used 2d feature extractors like SIFT, SURF, BRIEF and ORB in python's OpenCV library to extract muzzle features and Flann matcher for pattern matching.
Face Detection in Live stream- Created a face detector and recognizer from video stream using CNN model and OpenCV in python.
Fraud detection and investigation - A fraud detection and investigation project PoC on transaction data using
multiple platforms like bank transfer and UPI. Delivered SQL database queries for tagging suspicious activities on transaction data. Single handedly developed schema and solution for investigating fraudulent activities using
Neo4j's graph database on millions of records.
Incentive Management System- A platform for analyzing incentives for sales executives. Build quota
recommendation system using Random Forrest regressor in python for an incentive management project.
Created descriptive analytics and dynamic dashboards for What-If analysis using Tableau dashboards.
Sql, Python, Video, Database, Tableau, Management, Analytics, Sales, Test, Insurance, OpenCV, Live Stream, Feature, Neo4j, 2D
2014 - 2017
job
Data Analyst
EY GDS.
On-demand VM Provisioning - A platform to spun up VMs in Azure using ARM template and DSC scripts.
Developed PowerShell scripts for deployment of ARM Templates, Management of resources in Azure, Active
Directory group creation, Azure Active directory syncing etc. Worked in developing desired state configuration
scripts for software installation in Windows machines spun up in Azure.
KYC Mapping - A KYC mapping platform from multiple sources. Extensively worked on a Python project for
mapping entities of KYC data using heuristically built pattern matching solution for a big data project in banking
domain in a time efficient manner.
Expense Analytics - Analytics on reimbursement data for finding data discrepancies. Designed and implemented workflow for the data aggregation and analytics on Expense data using SQL for finding
discrepancies in expense reimbursement using a predefined rule engine and used Tableau to create dashboards
with summary page and drill down capabilities. Used SSIS modules to include PGP encryption to encrypt the data to send it across the server securely.
Social Media Text Analytics - A relevant news aggregator of CXO level changes for business executives. Built
a spam filter for filtering ad contents from the captured supervised data using Naïve Bayes Algorithm. Introduced
multiple NER taggers such as Nerd, CCG etc for classifying the data into clusters of unrelated events and removing duplicates.
Social Media Text Analytics PoC - A proof of concept for aggregating relevant news from multiple sources.
Designed workflow and implemented system using R and Python for capturing EY relevant up-to-date content from social media and deliver it after processing on demand. Incorporated multiple API's such as LinkedIn API,
Twitter API, Google news API, Wikipedia API etc to the system to enrich the data and provide deeper insights into
different entities.
Developed PowerShell scripts for deployment of ARM Templates, Management of resources in Azure, Active
Directory group creation, Azure Active directory syncing etc. Worked in developing desired state configuration
scripts for software installation in Windows machines spun up in Azure.
KYC Mapping - A KYC mapping platform from multiple sources. Extensively worked on a Python project for
mapping entities of KYC data using heuristically built pattern matching solution for a big data project in banking
domain in a time efficient manner.
Expense Analytics - Analytics on reimbursement data for finding data discrepancies. Designed and implemented workflow for the data aggregation and analytics on Expense data using SQL for finding
discrepancies in expense reimbursement using a predefined rule engine and used Tableau to create dashboards
with summary page and drill down capabilities. Used SSIS modules to include PGP encryption to encrypt the data to send it across the server securely.
Social Media Text Analytics - A relevant news aggregator of CXO level changes for business executives. Built
a spam filter for filtering ad contents from the captured supervised data using Naïve Bayes Algorithm. Introduced
multiple NER taggers such as Nerd, CCG etc for classifying the data into clusters of unrelated events and removing duplicates.
Social Media Text Analytics PoC - A proof of concept for aggregating relevant news from multiple sources.
Designed workflow and implemented system using R and Python for capturing EY relevant up-to-date content from social media and deliver it after processing on demand. Incorporated multiple API's such as LinkedIn API,
Twitter API, Google news API, Wikipedia API etc to the system to enrich the data and provide deeper insights into
different entities.
Content, Twitter api, UP, Social, Processing, KYC, Google, Server, Software, Analyst, Azure Active Directory, It, Twitter, Workflow, Ssis, Banking, Social Media, Windows, PowerShell, Analytics, Management, Linkedin, Tableau, R, Deployment, Active Directory, Big Data, Azure, API, Python, Sql
My education
National Institute of Technology
N/a, Computer Science Engineering
N/a, Computer Science Engineering
Jomy's reviews
Jomy has not received any reviews on Worksome.
Contact Jomy Sebastian
Worksome removes the expensive intermediaries and gives you direct contact with relevant talent.
Create a login and get the opportunity to write to Jomy directly in Worksome.
38000+ qualified freelancers
are ready to help you
Tell us what you need help with
and get specific bids from skilled talent in Denmark