Omidzamani

This skill optimizes DSPy programs using mini-batch Bayesian optimization with rich feedback signals to balance efficiency and performance.

This skill designs type-safe DSPy module signatures with input/output schemas and Pydantic models for robust, structured results.

This skill refines DSPy outputs through iterative reward-based checks and best-of-N selection to meet format, length, and content constraints.

This skill guides you through auditing and fixing a plugin skill file in one pass, delivering a verified, ready-to-report result.

This skill helps you build a retrieval-augmented generation pipeline using ColBERTv2 in DSPy to produce grounded, factual answers.

This skill helps you debug, trace, and monitor DSPy programs in development and production with inspection, MLflow tracing, and custom callbacks.

This skill helps you design production-ready DSPy custom modules with proper architecture, state management, serialization, and testing.

This skill distills a DSPy program into fine-tuned weights for efficient production deployment and reduced inference costs.

This skill helps you systematically evaluate DSPy programs with built-in and custom metrics, enabling baseline creation, comparisons, and production readiness.