Crowdsourcing infra for model inference: A conceptual exploration of building a distributed LLM inference network that crowdsources GPU capacity, rejects unsuitable network nodes, splits model layers across capable machines, and coordinates execution through pipeline parallelism.

Building a Key-Value store: A practical walkthrough of building an append-only key-value store, the problems that appear as it grows, and the design choices used to address them. Inspired by Designing Data-Intensive Applications by Martin Kleppmann and Michael Greim.

Iris - A Case Study: how I built applywithiris.com and the challenges I faced.