

Deep debugging and kernel-level contributions to ONNX Runtime


Extending CPU/GPU LLM inference kernels

Novel algorithm for single-source shortest path computation in directed graphs

Real-time data streaming platform for agricultural analytics

Optimized algorithms for maximum clique detection in large graphs

Efficient weighted matching algorithms for poly-streaming graph processing

Optimized graph-based indexing for approximate nearest neighbor search

Hybrid CPU-GPU framework for parallel depth-first search with cost constraints


High-performance real-time imaging system for Raspberry Pi


Macro-based OOP framework for C with classes, inheritance, virtual functions, iterators and extensive data-structure library.


High-performance key-value store with memory swapping capabilities


Model optimization and performance contributions

AI-powered optimization engine for next-generation computing