SaiShreya Nuguri

I’m a PhD Candidate in Computer Science at the University of Missouri - Columbia researching adaptive AI systems for immersive learning environments. I develop LLM-powered, RAG-based pedagogical agents integrated with reinforcement learning–driven adaptivity mechanisms to dynamically personalize VR cybersecurity training systems.

I designed and built a modular LLM/RAG pipeline (document ingestion → embedding → retrieval → prompt orchestration) that powers a real-time pedagogical agent inside a Unity-based VR cybersecurity training platform. I engineered RL-driven adaptivity mechanisms that adjust instructional difficulty and feedback cadence from live learner-behavior and physiological signals, measurably reducing cognitive overload in pilot studies. I also architected AWS-backed data and logging pipelines to capture, store, and analyze learning-performance telemetry at scale, supporting reproducible evaluation across participant cohorts.

During an Altair DevOps internship, I built and deployed a production RAG-based LLM assistant that ingests CI/CD and cloud infrastructure logs, performs semantic retrieval, and surfaces root-cause hypotheses—cutting mean debugging time for engineers. I led the end-to-end pipeline (document ingestion, embeddings, retrieval, and prompt orchestration) and integrated it directly into DevOps workflows to automate triage of ETL failures and platform misconfigurations. Earlier, I performed structured root-cause analysis of ETL job failures and performance bottlenecks at Informatica, and I designed and deployed secure RESTful APIs on Google Apigee at Stratus Meridian.

My work is anchored in secure, empirically evaluated systems: I architected a HIPAA-compliant cloud-based IoT framework for remote patient monitoring and conducted STRIDE-based threat modeling and risk assessment. I serve on the iLRN conference organizing committee and have received Best Academic Presentation (2024) and Rookie of the Year (2025) awards. I also lead NSF REU research teams, mentoring students and delivering AI/VR technical workshops, with a professional focus on adaptive learning that’s rigorous, secure, and measurable.