Provider-agnostic enterprise RAG and agent evaluation harness for Azure Foundry, vLLM, Ollama, and local demos.

Single-agent, evidence-grounded claim verification to catch LLM hallucinations — a pluggable fact-gate for agent-arena and any multi-agent system (CrewAI, AutoGen, LangGraph).

Evaluation patterns, release gates, and anti-hallucination techniques for developer-focused AI workflows.

TypeScript eval harness for measuring whether Grok answers stay grounded in source evidence

Detect & score LLM hallucinations by groundedness — labeled data, precision/recall/F1, runs offline with no API key. Pluggable LLM-judge backends.

RAG evaluation workbench for retrieval recall, citation coverage, groundedness checks, and failure analysis

A local groundedness judge for RAG: QLoRA-distilled to match a frontier judge 100% at $0/call. Ships only if its own evals beat baseline.

Deterministic citation and claim-support checks for RAG evaluation datasets.

Retrieval-Augmented Generation (RAG) research application with groundedness evaluation to increase reliability and confidence in LLM generated output (post-hoc hallucination detection).