8 projects
argus-llm
Heuristic quality scoring and monitoring for LLM outputs. G-ARVIS scores for Groundedness, Accuracy, Reliability, Variance, Inference Cost, and Safety.
keelgate
Safety-first agent harness for AI agents that touch money and regulated actions. The LLM proposes; deterministic code decides.
tycheon
Calibrated financial forecasting and risk. Kronos forecasts the path; Tycheon tells you how much to trust it.
provenir
The trust layer for model post-training: reproducibility, contamination-safe eval, reward-hacking detection, supply-chain scanning, TIM detection, retraction monitoring, LoRA lineage signing, Sigstore DSSE, CycloneDX/SPDX SBOM export, EU AI Act / FDA PCCP / NIST AI RMF regulatory evidence, and M&A acquisition packaging
agentmesh-proxy
Governance proxy for every AI tool — PII/PHI masking, prompt injection detection, anomaly alerts, semantic cache, token quotas, vendor routing, compliance reports
bulwark-agent-security
Enterprise-grade defense framework for AI agents — protects against prompt injection, data exfiltration, and memory contamination.
pytorchforge
Enterprise-grade PyTorch framework with governance, monitoring, and production deployment capabilities
pulseflow-mlops
Production-grade open source MLOps pipeline for enterprise data engineering and predictive modeling
