Devesh Tiwari
Papers
1
Total Citations
2
H-Index
1
About
Devesh Tiwari is a leading researcher in high-performance computing (HPC), distributed systems, and machine learning infrastructure, with a focus on optimizing resource efficiency for emerging workloads. His major contributions lie at the intersection of serverless computing and deep reinforcement learning (DRL), where he pioneered novel frameworks to accelerate training pipelines. His highly cited work, "Nitro: Boosting Distributed Reinforcement Learning with Serverless Computing" (2024), introduces a serverless architecture that dramatically reduces the overhead of DRL training—critical for applications in gaming AI, robotics, and system scheduling. This paper has already garnered significant attention for its practical impact on scaling trial-and-error learning processes. Tiwari’s research addresses fundamental bottlenecks in distributed systems, enabling faster and more cost-effective deployment of AI models. With a growing citation footprint, his work is shaping the next generation of cloud-native machine learning platforms. He is also recognized for advancing energy-efficient computing and fault-tolerance mechanisms in large-scale systems, making him a key figure in bridging the gap between theoretical AI advances and real-world HPC deployments.
Research Focus
Key Achievements
Top Papers
- 1