Optim Hire

Go-getter, passionate about technology, innovation. Eager to utilize technical skills combined with AI/ML domain knowledge to create practical products for the real world

Responsible for detailed EDA, model creation, deployment, testing of model, monitoring, and retraining approach. Achieved Recall of 82 using F2 score as metric.

Task was to notify our customers when new loan product is available for them Also at regular intervals track the gps coordinates using silent notifications Other use case was sending reminder notification for emi delay. Notification type was sms,in app,and calls .

Golang GRPC Microservices for Project Management System

Built robust grpc based services,Developed Micro services using Golanguage and developed corresponding test cases for Company researchNow.

GRPC GPS Logger Microservices
Built an logger service that can handle million writes of gps coordinates per 10sec,used channels extensively,Wrote code for logrotation ,log compression .API is being served as Grpc call via proto buffer

Task was to help indentify customers prior who are likely to default a 5K short term loan Product, Prior to model the default rate for this product was 10 % , for all our uber partner driver customers

Outcome: Calibrated ML model helped to drop Loan Default rate

Created Golang grcpc webservice that uses vault to run authentication service.

Wrote service that creates users,assign roles,create certificates,authenticate using certificate etc

Built Authentication services for the team that is building elastic load balancers. Worked as freelan

Worked on building rest,grpc based APIS for the product.

Build rest apis using spring mvc and hibernate for Mintwalk and supermoney both.

Further to add on also built microservices using GRPC,Golang .And used Docker to containarize the app.

Developed Timeband rules for users booking their events on BookMyShow,Developed system which recommended nearby venues,Also developed Recommendation system which used Movie Ratings to generate recommedations,Here we used User-User collaborative filtering,Movie-Movie similarity for recommendation,

Task was to automate KYC verification of customers Identifying eye descriptors and then calculated EAR I.e eye aspect ratio for consecutive frame to identify if person in frame is blinking his eyes or notBuilt face verication module to verify profile photos with document photos.Built name and add

Details: Built a highly scalable logger service using Golang that can handle a million writes per 10sec, used channels extensively, Wrote code for log rotation, log compression Deployment using Docker
Outcome: High performance Logger

Details: Created ML Recommendation engine for music, used Music Information Retrieval (MIR) from the library by Vienna university, used features like rhythm pattern, temporal descriptor, etc(approx 2000dimensions of one song) Used Locality Sensitive Hashing algorithm +Ep

Task was to predict likelihood of patient having heart disease,
Role,
Cleaning of data, feature engineering,choosing up performance metric(F2 score),
Model Selection,Hyperparameter tuning,Feature encoding.
Optimizing threshold using PR curve for

The Objective was to predict testing Time for various combinations of configurations.This could be useful to prioritize the configurations to be tested that could save most of our time. - Given a set of feature variables predict testing time.

Challenge was given various data like applicants- information about each loan application at Home Credit)bureau,we had to predict what is the application's capability to repay the loan

Built a machine learning model to predict whether two questions asked on quora are similar or not. So that similar questions asked may have the same answers which have been given earlier for the previously asked similar question

Used a pretrained Vgg16 model to extract visual features from outfit, extracted text features from descriptions ofoutt like like weighted tdfword2vec Used combination of above features to nd the immediate neighbours tried both using Euclidean distance,and also trained LSH model

Developed a CNN model for identifying document types like PAN,AADHAR FRONT,AADHAR BACK,CHEQUE,

Was responsible for model training ,creation and deployment.

Identified eye descriptors, calculated EAR for blinking detection, built face verification module, and similarity validation between provided names.

Performed data engineering, data pipeline, and detailed analysis of SuperMoney customers. Predicted first-time Uber Customer default rate with 0.79 F-Beta Score.

Predicted testing time for various configurations to prioritize testing and save time.

Built a machine learning model to predict similarity between two questions on Quora.

Predicted the application's capability to repay the loan with an AUC of 0.77.

Worked on backend,Worked on building rest apis, prediction models,worked on NoSQLs.

Primary languages worked on their was Php,Golang,C#,NodeJs

Database MSSQL, MongoDB,Neo4j,CouchBase,Redis

Worked on multiple projects ranging from building rest api services using Spring MVC hibernate,grpc microservices, graphQL too.

Also worked on SMS parsing projects involved natural language processing using nltk,spacy,

Also worked on building Machine learning models, for predictin

Notification System

Created Golang grcpc webservice that uses Google Cloud
Notification system
pub sub for creating
Wrote GRPC based webservice in golang and created processor which listens to
subscriber and sends notification to users.

Built backend powered with microservices using rest/grpc apis. Built RealTime Notification Engine, Calibrated ML model for loan default prediction, Computer Vision Module, and scalable distributed systems. Also built SMS profiling using regex.

Worked in core team building recommendation engine, backend services, Music Recommendation Engine, and real-time seat occupancy module. Also worked on cold start problem of Music Recommendation Engine and developed machine learning models.

Helped client Mercedes with building marketing dashboards in Qlikview.

Vinay S.

Vinay S.

Time zones ready to work

Signup to hire this candidate

Interview Videos

Portfolios

CardioVascular Disease Detection

Event Notification Engine

Golang GRPC Microservices for Project Management System

Golang GRPC GPS Logger Microservices

Predicting short term loan default

Authentication services

Mintwalk and SuperMoney Entire Backend API

Recommendation Engine based On History and Ratings (BigData)

SmartBackGround Verication using Deep Learning,NLP and Computer Vision

AscertLogger Microservices

Music Recommendation Engine(Bookmyshow Hackathon)

Cardio Vascular Disease Detection (Hackathon May2020 )

Predict Mercedes-Benz Greener Manufacturing Configuration Testing time (Kaggle) Top 1 %

HomeDefaultCredit Risk (Kaggle 0.77 AUC) top 30%

Quora Question Pair Similarity (Kaggle)

Fashion Outfit Recommendation based on Content Recommendation

Document Detection using CNN

SmartBackground Verification using Deep Learning, NLP, and Computer Vision

Short Term Loan Default Prediction

Predict Mercedes-Benz Greener Manufacturing Configuration Testing time

Quora Question Pair Similarity

Home Default Credit Risk

Employment

Bookmyshow

Backend Developer/Data Science/AI

Data Engineer Freelancer

Software Engineer Backend,AI/ML/NLP

Software Engineer

BI Consultant

Education

US work authorization

Skills

Hobbies

Achievements

Certification

Preferred Languages