AI Firewall and guardrails for LLM-based Elixir applications

Interactive Phoenix LiveView demonstrations of the Crucible Framework - showcasing ensemble voting, request hedging, statistical analysis, and more with mock LLMs

Data validation and quality library for ML pipelines in Elixir

Explainable AI (XAI) tools for the Crucible framework

Advanced telemetry collection and analysis for AI research

Structured causal reasoning chain logging for LLM transparency

Request hedging for tail latency reduction in distributed systems

Experimental research framework for running AI benchmarks at scale

CrucibleFramework: A scientific platform for LLM reliability research on the BEAM

Adversarial testing and robustness evaluation for the Crucible framework

Fairness and bias detection library for Elixir AI/ML systems

Dataset management and caching for AI research benchmarks

Statistical testing and analysis framework for AI research

Multi-model ensemble voting strategies for LLM reliability