Raghavasai Kowlasker

• Improved Pipeline Historical Data report visibility by including direction details, resulting in 50% decrease in application dependencies and increased data observability for insightful business metrics.
• Built consolidated summary of PHD report to inspect over 6000 miles of gas pipelines, increasing application performance by 60% and improving efficiency for field technicians to understand pipeline defects.
• Developed full stack web feature for corrosion compliance records application, transforming manual data entry into a one-click job and providing a single view of reports to avoid error-prone manual entry.
• Enhanced RESTful web service APIs by removing database dependencies, resulting in a decrease of data latency by 70%.
• Refactored gas pipeline main flow and segmentation according to client requirements, improving performance of ArcGIS tools application by 75%.
• Leveraged C# and MS SQL server for CRUD operations in gas reference pressure calculation and corrosion compliance applications.
• Utilized Azure Data Studio for backend integration work in an agile environment.
• Collaborated with cross-functional teams to enhance application based on client requirements.
• Worked with AngularJS and Python to build ArcGIS Map tools application.
• Streamlined application performance by improving code quality and reducing bugs.
• Demonstrated ability to work efficiently in a fast-paced environment while meeting tight deadlines

• Derived analysis leveraging Jupyter notebook to find valuable insights that helped recapitulate client’s code health metrics. Identifying the potential flaws with the help of the health metrics model reduced the risk of the metrics by 75%.
• Optimized API endpoints to a single notebook to analyze and predict health metric activity, resulting in a 50% increase in performance and a 30% increase in accuracy.
• Designed and architected a Proof-of-concept chatbot in the healthcare domain for improving pregnant women’s nutrition literacy. Integrated the user interface using natural language processing from Google Dialogue Flow and hosted the chatbot application in GCP, resulting in a 70% increase in user engagement and effectiveness.
• Reviewed journels series in ACM organization on Human Computer Interaction (HCI) is a journal series and acted as a 3AC reviewer, contributing to the advancement of HCI research and knowledge dissemination.

• Optimized project performance by designing and normalizing the Database model components to 3NF, resulting in a 30% reduction in query response time.
• Contributed to the project's scalability by successfully configuring Redis Cluster in the project lab and on AWS ECS Cluster, increasing the system's capacity by 50%.
• Improved project outcomes by transforming user requirements into 50+ user stories and tickets, following Agile methodologies and Continuous Integration using Azure Devops throughout the software development life cycle.
• Developed the front-end using Vue3, HTML, and Javascript, resulting in a 20% increase in user engagement and an improved user experience.
• Ensured high-quality deliverables by developing 100+ integration test cases and improving code coverage to 95%, resulting in a 40% reduction in bug density.

• Developed a new website for Marked by Covid organization from scratch with a team of 9, utilizing Python, Vue3, Django framework, and PostgreSQL.
• Created a central repository of Covid memorial monuments on the website, which has been recognized as a pandemic remembrance for lost ones due to Covid.
• Implemented a Google Maps API solution for memorial monuments and featured web pages, enabling users to submit, search, and acquire information about these events.
• Integrated a donation option on the platform for individuals looking to contribute to this meaningful cause.

A rental maintaince Application

• Developed and enhanced a web-based tool to streamline the management and monitoring of repair requests in a residential complex.
• Created a user-friendly platform that caters to four unique personas: property manager, building manager, tenants, and maintenance staff.
• Utilized Python, the Django framework, and Postgres database to construct a robust and efficient tool.
• Deployed the platform on Heroku for seamless accessibility and usage.
• Designed and implemented a solution to address the specific needs of a residential property management setting.
• Engineered a solution that simplifies and expedites the repair request process, improving tenant satisfaction and operational efficiency.
• Built a scalable and adaptable platform to manage repair requests and maintenance schedules for a residential complex.

• Developed a steganographic model using the Least Significant Bit (LSB) and Mist filter algorithm as a foundation, generating an encrypted key to secure data.
• Implemented a Neural Network model to enable encrypted data of any format, including text, image, or video.
• Utilized Python, JavaScript, CSS, and HTML to develop the model, which was deployed on PythonAnywhere.
• Contributed to the field of data security by creating an advanced steganographic model capable of encrypting various types of data using a Neural Network approach.
• Enhanced the OCD to design all the required servers for software development in large scale industry.

• Initiated estimation of drought severity using Standard Precipitation Index(SPI) and Palmer Drought Severity Index(PDSI) models, predicting index scale from -3 (highest drought) to +3 (fertile land).
• Utilized 100 years of Indian Meteorological Department dataset to train and test the model, employing Support Vector Machine, Artificial Neural Network, Naive Bayes, and Random Forest algorithms to determine the best accuracy.
• Achieved 97% accuracy from the Random Forest algorithm, utilizing Python, Jupyter Notebooks, and MS Excel to develop the model.
• Contributed to the field of drought prediction by applying advanced data analysis techniques and utilizing machine learning algorithms to accurately predict drought severity.

• Led a bachelor project team to design and develop a custom online store with 10 product categories and an inventory of over 40 items, from ideation to implementation.
• Implemented advanced data mining techniques such as content-based filtering and K-Nearest Neighbor algorithm to provide personalized product recommendations based on customers' past interactions with the website.
• Conducted thorough user experience testing and iteratively improved the website's functionality and interface, resulting in increased customer satisfaction and engagement.