Sreepallavi Manikonda

I've built AI systems spanning hybrid RAG retrieval, agentic reporting, computer vision, and topological analysis. My work combines Python, LLM pipelines, evaluation harnesses, and structured analytics.

At Flutura, I assisted in developing AI models for industrial pattern recognition and large-scale operational data analysis. At Main Flow, I automated data extraction and processing routines, reducing manual workflow overhead by 40%, and built Flask microservice endpoints for real-time distributed applications.

In my GraphRAG Hybrid RAG Pipeline, I engineered adaptive routing between vector search and graph traversal and benchmarked retrieval accuracy and failure modes. I also built a portfolio health reporting agent that turns messy multi-schema spreadsheets into executive-level monthly health summaries using a five-signal weighted RAG framework.

My research projects include an EEG phase-synchronization classifier that achieved 95.8% leave-one-out cross-validation accuracy and a task-conditioned point-cloud upsampling framework that improved uncertainty agreement from 0.23 to 0.54. I'm pursuing a B.Tech in Computer Science and Engineering with a focus on Artificial Intelligence.