prompt optimization

DriftWatch AI is a proxy and middleware layer for LLM traffic that trims redundant tokens, validates JSON schemas, and monitors model drift across OpenAI, Anthropic, and Gemini workflows.

A Python library that transpiles stateful imperative LLM workflows into optimizable DSPy programs, with typed outputs, self-correction loops, validation reporting, and prompt tuning support.

A GitHub release for Microsoft’s Agent Lightning Skill, which helps coding agents optimize other AI agents using benchmarks and iterative improvements across prompts, tools, workflows, models, and reasoning settings.

LEVI is a harness-first evolutionary framework for code and prompt optimization. It focuses on reducing LLM cost with diversity-preserving search, role-aware model routing, and a proxy benchmark, and presents comparative results against several existing systems.

Google Research paper describing a deployed multimodal defense system for detecting coordinated synthetic spam and abusive media at scale, using account relatedness, synthetic content classification, and LoRA-tuned LLMs.