failure-modes

ML Approaches for RUL Prediction, Anomaly Detection, Survival Analysis and Failure Classification

Physics-informed neural networks for highly compressible flows 🧠🌊

Collection of how and why software systems fail

RMC-TotalRisk is a powerful risk analysis software package designed to support dam and levee safety investment decisions.

[UAI 2023] Official implementation of Efficient Failure Pattern Identification of Predictive Algorithms

Detecting Failure Modes in Image Reconstructions with Interval Neural Network Uncertainty

Enabling Model-Based Diagnosis and Failure Model Generation with Active Automata Learning

Agent Failure Patterns - A field guide to how AI agents fail in production - documented failure patterns, root causes, and mitigations for teams building agentic systems.

Reproduce real backend failures on Spring Boot and Kafka with one command. Each scenario ships the symptom, the root cause, the fix, and a test that proves it.

PressureX is an engineering evaluation package for a passive layered structural mitigation concept using shear-thickening fluid behavior to broaden impulsive loads and reduce peak transmitted response in high-vibration aerospace structures. Targets are design-intent until validated.

Empirical benchmark comparing agent architectures (single-agent, multi-agent, adaptive) on ProgramDev-v0 and CyberGym tasks. Key finding: adaptive architecture > single-agent > fixed-pipeline multi-agent.

Source-available, measurement-first pulsed-energy testbed for tri-sector storage/discharge control, derated storage, phase authority, sensor-truth checks, energy accounting, kill criteria, evidence bundles, and human-reviewed scale-up gates.

Catalog of 300 distributed-systems failure modes in 17 categories — mechanism, impact, detection/mitigation — for design review, postmortems, chaos engineering.

Architecture for resilient, governed, and regenerative intelligence under uncertainty.

An identification key for 36 failure modes in AI-orchestrated agent systems — a field guide built from logged incidents on one instrumented fleet

WorldLag — Agent Failure Series #9. Interactive simulator of stale world model / belief-reality divergence in AI agents.

10 agent skills for the failure modes that don't throw: Windows + Git Bash quoting, exit-0 silent failures, encoding/BOM, git ref plumbing, network layering and REST fallback, silent data corruption, runtime resolution, agent sandbox boundaries, CDP, stale output layers.

How coding agents fail: 10 task specs, 3 executable fixtures, an F01-F13 failure taxonomy (F13 unscored in v0), grading rubric and redacted trace evidence. Disclosed examples, not an unseen holdout.

Runtime-agnostic hook harness that catches unverifiable prompts, enforces failure-mode templates, and gates task completion on passing tests.