Pravallika Daka

• extensive experience orchestrating AWS and Azure environments, architecting Snowflake data warehouses, and designing efficient data pipelines using PySpark and SQL.
• Proficient in the Hadoop ecosystem, including MapReduce, YARN, Apache Hive, Apache Spark, and Apache Kafka, for big data processing tasks.
• Adept at infrastructure automation with CloudFormation and experienced in Agile methodology.
• Skilled in data visualization with Tableau and proficient in optimizing queries and data analysis in Snowflake.

Food Inspection Data Analysis

• Utilized diverse graph theory methods, such as Depth First Search (DFS), Betweenness Centrality, Closeness Centrality, and Degree Distribution, to construct and analyze a network of over 3,000 food establishments, pinpointing high-risk locations and optimizing inspection strategies.

Failure Prediction using Ensembling Techniques

• Led the creation of predictive models for software failure anticipation, crafting resilient prediction frameworks that incorporated classifiers such as the XGB Classifier and Gradient Boosting, as well as ensemble methods like Soft Voting and Hard Voting. This pioneering work culminated in a research paper presented at ICMLAS in February 2022, showcasing our innovative approach to software failure prediction.

• Research paper published at ICMLAS in February 2022, for the project paper "Failure Prediction using Ensembling Techniques"