Insights • Slides • Presentations

Presentations and case studies from our work on specific projects.

BrassinAI case studies, product walkthroughs, and presentation-style explainers for applied AI systems and research-led engineering work.

Nano vLLM Scheduler

A presentation on nano-vLLM scheduler behavior and inference-serving mechanics.

INFERENCE

GPU Programming & LLM Serving: A Practical Primer

From silicon to serving: GPUs, the execution model, GPU DSLs and Triton, CUDA Graphs, FlashInfer attention, and vLLM.

HPC

Inference: Quantization and Attention workflow

A Brief introduction to Quantization and Attention workflow for Inference

INFERENCE

Devocle - RAG for DevOps Engineers

A case study in building a fine-tuned RAG system for DevOps log debugging, from dataset curation to Docker Desktop extension.

RAG