Parag Sonar

Worked on End to End Delivery of Data Science Projects Of Pattern Recognition, Statistical Modeling, Fraud Detection Which involved Data Extraction, EDA, Wrangling, Building Machine Learning Ensemble Models Like Random Forest, XGBoost LightgBM, SVM, DT, Hyperparameter Tuning. Worked on Product Segmentation Model :: Created end to end Data Pipeline of Product Segment through SQL Query. Worked on NLP Problem Such as Sentiment Analysis, Named Entity Recognition And Key Phrase Extraction (Text Analytics), Money Laundering, Non-Parametric Statistical Model, RNN, Chatbots. Work on Segmentation :: Used Transaction Level and Demographic Data to Build Custom Customer Profiles. An Ensemble of K-means and Business rules were Applied to Build the Segments using Python. Fraud Detection Analytics :: Identify Fraudulent Transaction Metrics by Building a Fraud Detection Model using Classification Models. Build an ANN Based Model for Predicting the Approval or Reject the Loan Applicants Cases. Client handled like (Shriram Finance, IIFL Finance)

Identify Valuable Data Sources and Automate Operational processes using Data Analytics & Machine Learning Techniques. Created Power - BI Dashboard for Bridging the Gap Between Technology and the Retail Business Using Data Analytics through Deliver Data-Driven Recommendations and Reports to Executives and Directors. Used SQL Query and Power Query for End to End Product Sales Data Pipeline. It Helps for Sales Prediction in Festive Seasons. Analyse Large Amounts of Sales Information using Python, R - Programme to Discover Trend Analysis and Patterns Recognition. Build Predictive Models and Machine-Learning Algorithms.

Implemented Automated Data reduction and evaluation with Map Reduce and Hadoop to reduce the process from eight (8) weeks to eight (8) hours. Created Data Visualisations and Data Crawling tool like Python and R Programming and for Visualization using Power - Bi and presented it Technology & Production Team. Developed and Deploy Advanced Statistical Models, Predicitve Models & Machine Learning Methods (Random Forrest Method, Decision Tree Method, SVM, NLP, Gradient Boosting, Supervised/Unsupervised Learning, Clustering, Classification and Regression Modelling).

Developed and Maintain End to End Database Using SQL Query and Data System in Reorganise and Readable format for Client like (PWC, Deloitte). Use Cluster analysis And Sentimental Analysis for Online Reviews for Quality Check with Product and Services. Use Statistical and Programming Tools like (SPSS, STATA, R - Programme, Excel, SQL, Python, and Power - Bi, Hadoop). Using Apache Hadoop creating investment Models and Trading algorithms. Worked on NLP Problem Such as Sentiment Analysis, Named Entity Recognition And Key Phrase Extraction (Text Analytics), Money Laundering, Nonparametric Statistical Model, RNN, Chatbot. Segmentation : Used transaction level and demographic data to build Custom Customer profiles. An ensemble of K-means and Business rules applied to build the segments using python.

Help clients to make decisions based on facts & data driven solutions using AI/ML Service offered are drug discovery, clinical trails predictions, biomarkers, sentiment analysis etc. Saved around 70% resource time from Planning and Execution. Saved around 80% resource time using Data Visualization tools like Power - Bi, MS - Excel, Python, R Programming, Hadoop in Pharma Domain for Data Forecasting. Worked to Design an Advanced Neural Network based Model that can Predict the Probability of Success of Clinical trials with Precision of 85%. Increased Data Scraping Capabilities & Data Quality by 4X. Raised Code Quality through Data reviews that helps to reduce deployment problems by 60%. Developed an Internal Web - Based Solution Database, Bringing team resolution rates from 60% to 90% in 3 Months. Created the Power - Bi Dashboard for Client like Commerzbank Group, Pfizer, Ranbaxy. Used Python and SQL for Time Series Analysis and Cluster Analysis. Build and Maintain Database Pipeline of Heart Rate Data and SQL Queries to track the operational Productivity.

Initiated the implementation of Scrum Methodology, resulting in 40% increase in Quality and Productivity. Perform Pharma Clinical Trails Data collection and a variety of Statistical Analysis using R - Programming, SQL, Python and Tableau, Power - Bi. Develop and Deliver Customized Solution for the biggest banks and Finance Companies in more than 6 countries. Worked Closely with a team of Data Engineers and Data Scientist to improve the efficiency Clinical Trials Prediction Engine by 67%. On Daily basis 40+ bugs were fixed. Used Microsoft Azure Health Data Services for Patient Heart Rate Data for Sensor Detection in Product Development using Azure Databricks.

Performed Data Crawling and Coding for 2 Projects. Maintain 99.5% Quality Level in Internship Period. Developed and Maintained 15+ Financial Models for Financial Institution.

Economics Major and Minor in Specialisation in Agribusiness Economics.

Costing Major with Specialisation in Mathematical Finance.