SUMMARY:

• Over 7+ years of experience building scalable SaaS products and solving practical business problems for startups to fortune 500 clients.
• Experience in creating Data Pipelines, performing Web Data Mining, Data Extraction, Data Transformation, Data Cleaning, Data Modelling, Data Wrangling, Statistical Modeling, Machine Learning, Data Visualization and Analytics.
• Well versed with latest technological demands, adaptability needs and data architecture.
• Good understanding of Data Models - various corporate multilevel schema design and implementation from scratch for data products and self contained implementations- experience in moving messy data structure (in TBs) to structured ones.
• Well experienced in Web Mining / Data ETL (over half million websites), extracting raw content in distributed fashion, pipelining to data lakes, applying Machine Learning on Scale and helping other data scientists/ ML engineers get relevant data for ML modeling.
• Experience in building synchronous/ asynchronous and distributed architecture on cloud infrastructure from scratch to reduce cost and time.
• Experience in Google APIs - geocode, translate etc. and well versed with designing best algorithms to use paid API’s efficiently and profitably.
• Experience in most of cloud platform - AWS, GCP and Azure, also experience in exploiting private on-premise cloud services.
• Experience writing production level data pipelining and modeling code with unit testing and checks for fault tolerant, secure and scalable systems.
• Experience modelling, writing scalable ML algorithm implementations, Machine Learning models and recommendation systems.
• Experience in reproducing research and bringing latest technological enhancements to practice. Active participant in research projects and publications.
• Extensive experience in relational as well as non relational databases - NoSQL, MySQL 5x-8x, MongoDB, Cassandra, PostgreSQL etc.
• Experience maintaining servers, keeping track of logs, errors and faults, security and improvements and helping SDE’s in ELK data flow.
• Experience in AWS with provisioning and maintaining AWS resources such as EC2, EMR, S3, RDS etc.
• Good Knowledge of Data Warehouse Architecture and various schemas like Star Schema, Snowflake Schema.
• Experienced in Data Analysis - business presentation ready reports creation, proficient in gathering business requirements and handling requirements management.
• Experience in Big Data Technologies - Hadoop, HDFS, Hive, MapReduce, PySpark etc.
• Experience in BI/ visualization tools like Tableau, Plotly etc.
• Experience in version control - Github.
• Have good communication skills and believe in collaborative work,
• Experienced in working independently as well as in team.
• Experience communicator with clients on data productization requirements.

TECHNICAL SKILLS:

Programming Languages: Python, SQL, Java, Hive, R, Py: Spark, C, C++

Internet Technologies: JavaScript, Chart.js, D3.js, HTML5, CSS3, PHP, Bootstrap, Angular, Rest API s

Databases: MySQL, MongoDB, Cassandra, PostgreSQL

IDEs/ Development tools: Jupyter Notebook, Springboot, Tableau, POSTMAN, IntelliJ, Eclipse- Java EE, GitHub, MongoDB Compass

Platform: OSX, Linux, Ubuntu, Windows

PROFESSIONAL EXPERIENCE:

Confidential, Princeton, NJ

Data Science Engineer

Technologies used: Python, Beautiful Soup, REST Web Services, S3, Big Data technologies, EC2, EMR AWS, MS Azure, Flask, MySQL, Chart.js, D3.js, Java, Spring JDBC, Boto-S3, Java, HTML, CSS, AngularJS, Salesforce APIs, Plotly, Tableau, Jupyter, Postman, IntelliJ etc.

Responsibilities:

• Asynchronous text scraping thousands of websites
• Implemented parallelized data processing operations using Dask framework to clean and filter text data
• Implemented ML algorithms to extract accurate needed informations on scale.
• Designing and developed optimal API call algorithms on Geocoding and translation Google API services to produce readable english results with minimal cost and time.
• Performed contact sourcing ML based optimizers to retrieve client focused required results and tagging searches.

Technologies Used: Python, Async.io, Dask, BeautifulSoup, Requests, Json, selenium, scrapy, matplotlib, pandas, AWS, MongoDB, XGBoost, NLP, NER, Py-spark

Confidential

Responsibilities:

• Creating structured data pipeline with 40+ integrations of various data sources to filter, transform and validate the inflow of raw data.
• Performed Data Cleaning and Preprocessing, transformations and performing predictive modelling.
• Targeted analysis of sales and customer acquisitions.
• Target was to find key insights and opportunities designated to leverage the data intelligently, thus improving customer targeting and over data value to increase sales.
• Performed RFM analysis, customer-churn predictions, recommendation system, association rule mining, data enrichment and quality improvement.

Technologies used: Python, GraphLab, numpy, pandas, scikit-learn, tensorflow, keras, Tableau, Chart.js, D3.js

Confidential

Responsibilities:

• Developed robust machine learning models for cryptocurrency direction movement.
• Instrumental in creating infrastructure for complete pipeline for the project.
• Provided framework for identifying key features for stacked models.
• Identified key features for direction movement useful for day traders.

Technologies used: Generative & Discriminative Models, Python, MongoDB, Neural Network, Bitcoin, Quandl

Confidential, Bridgewater, NJ

Software Engineer - Data Analytics

Responsibilities:

• Design and Implement scaled productization algorithms for 4G wireless systems using advanced Self organizing network and machine learning techniques with Python, Matlab
• Improved the accuracy and computation efficacy of network
• Feature extraction, selection, analysis and optimization of algorithms using Python, MATLAB
• Apply machine learning/ reinforcement learning algorithms to large dataset, utilizing GPUs to accelerate training Processing
• Worked on creating Data Pipelines, strategizing and implementing Micro-Service based data infrastructure.
• Managed Cloud architecture, ensuring efficient data management and data governance.
• Wrote robust Machine Learning models to learn and solve practical business problems.
• Proactively improving and maintaining data quality and identifying data issues.
• Worked on REST API’s, Scraping and Crawling large web data, building Scalable SaaS products.
• Researched on recommendation engines to optimize the quality of algorithms used.
• Optimized ETL process for query efficiency and quality.
• Performed Data Cleaning, data pre-processing, visualizations and implementing on data pipeline.
• Worked on Big Data technologies - Hadoop ecosystem to ensure high performance on larger datasets.
• Worked on data maintenance in a logical, consistent, accurate and sustainable form.
• Worked on creation and analysis of data trend reports.
• Scraped and crawled web data from multiple sources and API’s to store in cloud data warehouse.
• Scaled and created non-relational NoSQL databases for efficient data ingestion on MongoDB servers.
• Optimized query efficiency on MySQL server and Hive on Hadoop ecosystem.
• Identified business logic required to clean, normalize and model incongruent source data.
• Performed descriptive and inferential statistical analysis of business data to find outliers and trends.
• Created dashboards on Tableau, D3.js, Chart.js and Plotly for user friendly interactive visualizations.

Confidential

Responsibilities:

• Exploiting TB’s of textual data from various magazines and news articles.
• Strategizing tagging and implementing solutions for model implementation on scale.
• Implemented Multi Class- classification algorithms of documents (in millions) into categories and reinspecting manual labelled documents to reduce human resource costs.
• Scaled implementable units of algorithms for tagging, increased precision by 12% and recall by 15% through pipelined structure of algorithms.
• Developed Flask Rest services to put tagged data into MongoDB and access sharepoint UI.
• Developed data access points, helped create SDE’s ELK dashboard.

Technologies used: Hadoop, EMR, Spark, Multithreading, Multiprocessing, Asynchronous programming, Natural Language Programming, Elastic Search, NER, Dask, Python, Flask, MongoDB, AWS S3, Redis, RabbitMQ