Hitesh Santwani

Software Engineer

McQuilling Partners•  June 2019 - Present

As an Intern in Information & Communications Technology Department I Worked Directly with the CTO on two Key Projects: Data Subscription Platform and REST API: The firm has collected vast amounts of Data on marine logistics data assets including vessel satellite data, commercial vessel data, and cargo flow data. ICT Department has been brainstorming and developing the roadmap for B2B Data Subscription Platform. This would create a new revenue stream Developing and Maintain the Data Pipelines for optimal Extraction, Transformation and Loading of Data into OLAP and subsequently into the OLTP. Developing the Machine Learning Models and Transforming them into the Production Ready State so they can be consumed by the firm’s Analysts and Traders around the world. Developing and Exposing the Restful API as a gateway for a subscription platform for the transport of data to the clients. 
Tech Stack: ASP.Net MVC5, C#, Visual Studio 2017, POSTMAN, AZURE (Data Factory, Azure Data bricks, SQL Data Warehouse), Git lab, T-SQL. Slack Intelligent Bot development: There have been constant efforts from the ICT Department to Increase the Productivity of the Traders and Ship Brokers on the floor. One of the ways it is achieved is by automating the manual workflows. The Slack Messaging client is the primary mode of the internal chat channel (for both formal and informal context). The intelligent bot is the Microservice which analyses the slack messages in real-time and extracts the business context to trigger the manual workflows. This Microservice communicates with other microservices for Machine Learning Capabilities (NLP and Classification). I designed and developed these microservices using ASP.Net MVC5 and flask along with Machine Learning models using python. Developed the Data pipeline to stream the thousands of Slack messages into the system.

Software Engineer

accenture•  June 2017 - July 2018

⇨ Client: JPMorgan Chase ⇨ Project: Global Portfolio Swaps - Custody Fund Services (Data Engineering Team) GPS - CFS is the cornerstone of the client’s Credit Default Swap Business. The Data Engineering Team is responsible for the enabling the Business Intelligence Team, Business Analysts, Risk Analysts, and Senior Management by serving the High-Quality Data, Inference and Prediction services, and Visualization Tools to make the Decisions. Our Team at Accenture Collaborated with Business Intelligence Team from London and Singapore to understand their requirements for new Data Sources, Prediction Capabilities, Custom BI Tools and Schema for new Data Sources. Our Team was responsible for developing the Data pipelines on Cloud to process the Terabytes of data from Data Vendors and Web Scrapped Data, to develop the microservices that would serve the Machine Learning models as service, to develop the RESTful API and Dashboards for the Business Intelligence, Business Analysts and other Business Stakeholders. The Skills Acquired and Exposure includes Software Development, Extraction Transformation and Loading, Machine Learning and Business Intelligence. Languages: Java and Python. Technologies: Hadoop, Spark ML, Spark SQL. Machine Learning: Logistic Regression, Ensemble Learning, Deep learning, Kernel Methods, Clustering Models, Classification and Regression Models. Cloud: AWS - EC2, ECS, S3, AWS Glue, AWS EMR, AWS Kinesis, AWS Sage Maker, AWS Comprehend, Redshift.

Associate Analyst

accenture•  October 2015 - May 2017

⇨ Client: JPMorgan Chase ⇨ Project: Derivatives Core Processing Platform. The Core processing platform is responsible for processing Over the counter derivatives and handles the diamond clients of the JP Morgan Chase. My Role in this project was of Java Software Developer working on the Spring MVC, Spring Core, Spring Boot and the typical Enterprise Java application stack (JIRA, Java 1.8, Eclipse, Jenkins, PL/SQL, Spring, Hibernate). The other major part of this project was modernization and automation of the manual workflows using Machine Learning and Big Data technologies. Our Team at Accenture had to present the incremental solution using prototyped REST Enabled microservices that would serve the prediction as to the service for the categorization of the trades. This enabled me to gain the skills and understanding of the process of converting the ML models to the Scalable and Production Environment Ready Services. The Data To train the models was streamed from Third-party market indicator providers and the web crawl of the Economic and Financial News Websites. The technologies used for the data Pipelines and ETL were MapReduce, Yarn, HDFS, Hive, Spark SQL, Presto, Pig, HBase, Parquet, continuous deployment tools (Jenkins), container orchestration tools (Kubernetes), Unix/Linux shell scripting. ⇨ Client: JPMorgan Chase ⇨ Project: Initial Talent Pool. In collaboration with Tech Lead and Project Management Employees as my client, we Developed Resource Time Sheet Analysis Automation Web Application. This Involved Developing a Full Stack system through which the clients can interact in a Distributed Environment. The Primary Function of this system was to parse the excel sheets (Time Sheets) and analyze data from the excel sheets for Issues, Reporting of aggregate Statistics, Cross-Validation with the JP Morgan Time Sheets (Contract Time Reporting System), and Automated Emails to the stakeholders. Stack - Angular, Spring Boot, Python (Pandas, Numpy, SciPy)

State University of New York at Buffalo, Buffalo

Data Science, MS•  August 2018 - Present

Master of Science in Engineering Science is a highly-selective program for students with a strong background in mathematics, computer science, and applied statistics. At UB the program focuses on developing Data Science Expertise in students by using the bottom-up approach in which core knowledge and skills in Linear Algebra and Probability are developed first followed by in-depth focus on Algorithms, Data Engineering and Machine Learning.