Search

DeepSeek Harness preset that brings fine-grained LLM verification, probabilistic pivot-tournament best-of-N selection, and checkpoint progress tracking into agent sessions.

Game-theoretic LLM training: Magnetic Mirror Descent, Quantal Response Equilibrium, Deep Equilibrium Models, token auctions, and magnetic optimizers.

LiteLLM's Auto Router v2, running inside the pi coding agent

Statistically defensible go/no-go verdicts for LLM model migrations.

LangChain adapter for textopt

LangChain tool for executing untrusted code in Microsoft Hyperlight micro VMs (experimental, AI-generated)

Deterministic runtime evidence and diagnostics for LangChain TypeScript agents.

Security testing and authorization-aware retrieval for production RAG systems.

Boruna: a deterministic, capability-safe programming language and framework for LLM-native applications