Reliability-first AI infrastructure on Elixir/BEAM. We build systems that make LLMs behave predictably in production.
40 public repos · 17 stars
Open research platform targeting 99%+ LLM reliability through ensembles, hedging, and statistical testing.
| Repository | Description |
|---|---|
| crucible_adversary | Adversarial testing and robustness evaluation for the Crucible fram... |
| crucible_bench | Statistical testing and analysis framework for AI research |
| crucible_datasets | Dataset management and caching for AI research benchmarks |
| crucible_ensemble | Multi-model ensemble voting strategies for LLM reliability |
| crucible_examples | Interactive Phoenix LiveView demonstrations of the Crucible Framewo... |
| crucible_framework | CrucibleFramework: A scientific platform for LLM reliability resear... |
| crucible_harness | Experimental research framework for running AI benchmarks at scale |
| crucible_hedging | Request hedging for tail latency reduction in distributed systems |
| crucible_ir | Intermediate Representation for the Crucible ML reliability ecosystem |
| crucible_telemetry | Advanced telemetry collection and analysis for AI research |
| crucible_trace | Structured causal reasoning chain logging for LLM transparency |
| crucible_ui | Phoenix LiveView dashboard for the Crucible ML reliability stack |
| crucible_xai | Explainable AI (XAI) tools for the Crucible framework |
Human-in-the-loop labeling infrastructure for ML datasets with inter-rater reliability.
| Repository | Description |
|---|---|
| anvil | Labeling queue library for managing human labeling workflows |
| forge | Sample factory library for generating, transforming, and computing ... |
| ingot | Phoenix LiveView interface for sample generation and human labeling... |
| labeling_ir | Shared IR structs for the North Shore labeling stack (Forge/Anvil/I... |
| Repository | Description |
|---|---|
| ExDataCheck | Data validation and quality library for ML pipelines in Elixir |
| ExFairness | Fairness and bias detection library for Elixir AI/ML systems |
| LlmGuard | AI Firewall and guardrails for LLM-based Elixir applications |
| datasets_ex | Dataset management library for ML experiments—loaders for SciFact, ... |
| eval_ex | Model evaluation harness for standardized benchmarking—comprehensiv... |
| hf_datasets_ex | HuggingFace Datasets for Elixir - A native Elixir port of the popul... |
| metrics_ex | Metrics aggregation and alerting for ML experiments—multi-backend e... |
| Repository | Description |
|---|---|
| hf_hub_ex | Elixir client for HuggingFace Hub—dataset/model metadata, file down... |
| nsai_gateway | Unified API gateway for the NSAI ecosystem—authentication (JWT, API... |
| nsai_registry | Service discovery and registry for the NSAI ecosystem—distributed r... |
| pilot | Interactive CLI and REPL for the NSAI ecosystem—unified interface t... |
| tiktoken_ex | Pure Elixir TikToken-style byte-level BPE tokenizer (Kimi K2 compat... |
| tinkerer | Chiral Narrative Synthesis workspace for Thinker/Tinker LoRA pipeli... |
| tinkex | Elixir SDK for the Tinker ML platform—LoRA training, sampling, and ... |
| Repository | Description |
|---|---|
| cns | Chiral Narrative Synthesis - Dialectical reasoning framework for au... |
| cns_ui | Phoenix LiveView interface for CNS dialectical reasoning experiments |
| Repository | Description |
|---|---|
| chz_ex | Elixir port of OpenAI's chz library - a powerful configuration mana... |
| embed_ex | Vector embeddings service for Elixir—multi-provider support (OpenAI... |
| ex_topology | Pure Elixir library for graph topology, TDA, and computational topo... |
| nx_penalties | Composable regularization penalties for Elixir Nx. L1/L2/Elastic Ne... |
BEAM Native · OTP supervision, telemetry, distributed resilience Research Backed · Every feature tied to reliability research Production Ready · Hex packages, docs, test suites
Updated 2025-12-24