Suma Katabattuni Email & Phone Number

College Station, TX, US

College Station, TX, US

Ingelheim Am Rhein, Rhineland-Palatinate, DE

• Developed a benchmark to replicate chemical reactions using 3 decision variables and 2 objectives, integrating MLflow forexperiment tracking and applied Bayesian optimization techniques, achieving a 10% yield.
• Designed and maintained an automation pipeline using Airflow to extract data from tables in Word documents stored in GCP,utilizing OCR techniques for text recognition, and saving it to a Microsoft Access database using.
• Implemented a GUI to visualize reaction parameters and performance of the reaction using ReactJs, Flask, and Python, anddeployed it on GCP, enhancing user interaction and data accessibility.
• Visualized optimized results from different single-objective optimization techniques using Tkinter and Matplotlib, aiding in theunderstanding of trends and the best-performing optimization technique.
• Created dashboards with Tableau to present results to scientists, facilitating data-driven decision-making and strategic planning.

College Station, TX, US

• Worked with a PHD student to support her research: To study the effect of nature on human stress levels.
• Designed webapp using OTree(python) framework along with HTML,CSS and Javascript with 6 tasks and conducted experiments with 256 participants over 32 sessions.
• Captured eye movements and data from the tasks using iomotions and OTree database.
• Cleaned the data using python and manipulated data into a consistent format to carry-out further analysis.

Seattle, WA, US

• Led the development of an end-to-end workflow management system, taking complete responsibility for its requirement gathering, design, implementation, testing and deployment and client feedback.
• Designed a GUI using ReactJS to help scientists visualize live status of a machine learning model workflow (GMFlow).
• Developed a wrapper script that eliminated the need for CLI commands and in-depth knowledge of GMLFlow, improving the user experience and making the process of model training more efficient and accessible for scientists.
• Designed and implemented RESTful APIs using Flask to establish communication with the backend and retrieve logs generated by GMFlow using PIPE and subprocess to make the troubleshoot process more efficient.
• Retrieved live status of the GMFlow by Implementing a file watcher to track updates to the stages and streamed the data to update the DAG displayed on the UI using the DOM structure to improve the overall user.
• Implemented comprehensive testing strategies to achieve high test coverage of 96% for with Jest tests for react components and 98% for Python unit tests, ensuring the reliability and robustness of the application.

Bangalore, Karnataka, IN

• Engineered SQL data transformations and decision tables in Cassandra, integrating Apache Spark and Databricks for scalabledata operations, enhancing personalized ads workflows with efficient data management.
• Developed a Random Forest classifier for customer segmentation, monitored machine learning models using MLflow,integrating Prometheus for performance monitoring and Grafana for visualization.
• Implemented targeted advertisement strategies based on customer segmentation and delivered personalized ads through variouschannels to enhance customer engagement and significantly improve conversion rates, resulting.

Mcs, Computer Science

Bachelor'S Degree, Electrical, Electronics And Communications Engineering

Intermediate

10Th