Quantizing LLMs for Low-Resource HPC CPUs: Lessons from HiPeC
How we achieved 87.25% accuracy on HPCorpus using ONNX Runtime, Grid-Ada LoRA fine-tuning, and GGUF compression on CPU-only cluster nodes.
Writing Custom CUDA Kernels for Real-Time Parallel Video Filters
Deep dive into 2D shared memory tiling, thread divergence mitigation, and CUDA stream pipelining for high-throughput video processing.
Zero-Knowledge Storage: Implementing XChaCha20-Poly1305 in Rust & Tauri
Designing client-side authenticated encryption, key derivation, and ephemeral memory wiping for KryptVault.