"CHINTHAN CHINNAPPA P T team on AMD AI League Hackathon"
Team Idea
We are building an adaptive AI inference engine that automatically chooses the most efficient way to run an AI workload based on the current system conditions and application requirements. Instead of using the same model configuration every time, the engine analyzes factors such as workload size, latency requirements, memory usage, compute utilization, and model precision, then dynamically selects the appropriate execution strategy. It can decide when to use different compute resources, batch sizes, model variants, or numerical precision to balance speed, resource usage, and inference quality. The system continuously monitors the results and learns from previous workloads to improve future decisions. Our goal is to make AI inference more efficient, responsive, and practical for real-world applications where compute resources are limited or workloads constantly change.
Team Members
.png&w=256&q=75)
Chinthan Chinnappa P T
chinthan_chinnappa
Submission
Submission draft in progress