Scholar
Amey Agrawal
Google Scholar ID: CXqdg3oAAAAJ
PhD Student at Georgia Tech
Citations & Impact
Citations
685
H-index
9
i10-index
8
Publications
14
Co-authors
22
list available
Publications
Resume (English only)
Academic Achievements
- “Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve” – OSDI'24
- “Vidur: A Large Scale Simulation Framework For LLM Inference” – MLSys'24
- “DynaQuant: Compressing Deep Learning Training Checkpoints via Dynamic Quantization” – SoCC'24
- “Maya: Optimizing Deep Learning Training Workloads using Emulated Virtual Accelerators” – to appear at EuroSys'26
- “Metron: Holistic Performance Evaluation Framework for LLM Inference Systems”
- “On Evaluating Performance of LLM Inference Serving Systems”
- “Sarathi: Efficient LLM Inference by Piggybacking Decodes with Chunked Prefills”
- “Learning Digital Circuits: A Journey Through Weight Invariant Self-Pruning Neural Networks” – NeurIPS 2019 Workshop, SNN 2021 Workshop
- “Delog: A Privacy Preserving Log Filtering Framework for Online Compute Platforms” – IEEE Big Data 2019
- “Logan: A Distributed Online Log Parser” – ICDE 2019
- Contributed to Singularity: Planet-scale, preemptible and elastic scheduling system for AI workloads at Microsoft
Research Experience
- 2025: Project Lead @ Project Vajra
- 2025: Research Intern @ Microsoft Research (with Sadjad Fouladi & Ganesh Ananthanarayanan)
- 2024: Research Intern @ Azure Systems Research (with Esha Choukse)
- 2023: Research Intern @ MSR Research India (with Ram Ramjee & Bhargav Gulavani)
- 2021: Research Engineer II @ Microsoft Research (with Muthian Sivathanu)
- 2018: Member of Technical Staff II @ Qubole
