Krishna P. Gudipaty

Research Summary

My research focuses on the challenges of operating ML/AI models under limited memory, compute, and network resources in edge and IoT environments. More specifically, I develop systems and techniques for fault-tolerant model serving, resource-aware model placement and scheduling, and efficient adaptation and serving of foundation and multimodal models. Much of my research asks how model design and systems design can be considered jointly rather than independently, particularly when failures, resource constraints, and changing workloads are first-class concerns.

CV: Download (PDF)

Publications

  1. Gudipaty, Krishna Praneet, W. A. Hanafy, N. Ng, et al., “HYDRA: Multi-level ensemble learning for failure-resilient edge inference,” in ACM/IEEE Symposium on Edge Computing (SEC), 2026

  2. Gudipaty, Krishna Praneet, W. A. Hanafy, T. Abdelzaher, et al., “Foundation models for the edge: Opportunities and trade-offs,” in IEEE MILCOM Internet of Things for Adversarial Environments (IoTAE) Workshop, 2026

  3. Gudipaty, Krishna Praneet, W. A. Hanafy, L. Wu, et al., “Practical considerations for failure resilient ml systems at the edge,” in IEEE MILCOM Internet of Things for Adversarial Environments (IoTAE) Workshop, 2025

  4. Gudipaty, Krishna Praneet, W. A. Hanafy, K. Ozkara, et al., “Mel: Multi-level ensemble learning for resource-constrained environments,” arXiv preprint arXiv:2506.20094, 2025

  5. A. Micciche, Gudipaty, Krishna Praneet, and S. Krastanov, “Quantum ldpc error correcting codes for use on 1d quantum dot arrays,” in APS March Meeting Abstracts, vol. 2024, S46–010

Website source: link, under GNU General Public License v3