InstructKR

A single-file open-source manifesto that borrows Agile’s structure, treats cognitive bandwidth as the new unit of work, and reframes AI-native development as orchestration, not code entry.

LogicKor turned Korean LLM evaluation into a test of grammar, instruction following, and cultural precision, then froze itself when the top models started crowding the ceiling.

An async Python framework that routes the same service through memory, Redis, or mocks at runtime, then invalidates only the keys that match.

A Python shell, a baked-in model layer, and a health-gated startup loop turn rvLLM into a serverless endpoint without dragging in a heavy serving stack.

A lightweight evaluation harness that shows how BM25, Korean tokenization, and a two-stage funnel shape what rerankers can actually prove.

A bare GitHub project can still reveal a lot. This one points to the way open-source AI now begins as a name, a license, and a bet on a workflow.

A tiny Python package turns capability transfer into selective weight surgery, so a model can import a new behavior without flattening what it already knows.

A Python pipeline that turns noisy AI posts on X into structured GeekNews-style reading, then keeps the whole system maintainable with async workers, dependency injection, and a disciplined data model.

A Rust implementation of Bayesian BM25 that calibrates lexical relevance into probabilities, then fuses sparse and dense signals with log-odds instead of guesswork.

How a localized LLM leaderboard uses manual stream parsing and a mathematical "underdog bias" to solve the Elo cold-start problem.

How a masterfully crafted .gitignore file reveals the exact tooling, philosophy, and future of a modern machine learning collaborative environment.

How a Rust-based CLI uses the Model Context Protocol and autonomous agents to write, test, and maintain its own repository.

How an accidental source code leak spawned an open-source AI agent framework that is now autonomously rewriting its own core in Rust.