A plugin for your agentic framework that optimizes code using the GEPA algorithm (Genetic-Pareto LLM-driven search).

Claude Code for DSPy: Comprehensive CLI to Optimize Your DSPy Code. our AI-Powered DSPy Development Assistant

CLI text optimizer built on GEPA. Uses Agentic Coding CLI's as mutator and observer -- no api keys required

GEPAzilla: open-source GEPA prompt optimizer with datasets, scorers, and telemetry.

Self-evolve Gemini CLI instructions, commands, and skills via the gemini CLI itself — GA + GEPA/DSPy, with hard gates before apply.

A benchmark, alignment pipeline, and LLM-as-a-Judge for evaluating the clinical impact of ASR errors.

A brief experiment applying GEPA (optimize_anything) to automatically compress Python solutions on code.golf.

Budge is the experimentation platform for agents.

A reproducible framework that uses DSPy + GEPA-style optimization to auto-tune LLM prompting pipelines for BI workflows (SQL generation, KPI summaries, executive status packs). Optimizes multiple objectives simultaneously—answer quality/faithfulness, token cost, and runtime—using offline eval sets, structured scoring, and Pareto-front selection.

Agent evolution lab: evolve autonomous-agent skills, tools, prompts, datasets, and evaluation loops from real usage evidence

Acoustic Semantic Instruction Register — LLM-guided hearing-device scene understanding and adaptive DSP parameter generation (DSPy + GEPA)

Reusable GEPA optimizer framework for typed candidate generation, evaluation, persistence posture, tracing contracts, and deterministic prompt optimization infrastructure.

Deterministic GEPA buildout examples and domain task fixtures for framework validation, redacted projections, evaluation traces, and local-first AI infrastructure.

Summit-Sim uses human-in-the-loop review to generate curriculum-informed, interactive backcountry emergencies for dynamic Wilderness First Responder (WFR) training.

Three-layer agent composition: DSPy signatures (declarative) + PydanticAI (runtime) + pydantic-graph (typed state machine) with GEPA hook via Agent.override()

kinn — a Bayesian diagnostic interview engine. Built with Opus 4.7 hackathon submission, Apr 21–28 2026.

Applied AI & Python Engineer

Evolvable AI programs: define a Python function + criteria, let an LLM iteratively improve it

DSPy/GEPA-powered self-improvement plugin for Hermes Agent skills, memory, and evaluator prompts.