Resilient AI Research

Research

We build resilient agent systems: bounded-state control, biologically-inspired coping signals, and practical evaluation under stress.

Resilient-GNN-Controller

A resilience-oriented controller prototype for AI agents. It maintains a bounded internal state ("hormones"), composes those into safe control knobs, and applies layered coping (timeouts, circuit breakers, adaptive baselines) to stabilize behavior under chaos.

Bench highlights

  • Preliminary train steps50k (per config)
  • Environments tested4
  • Evaluation100 episodes × 1000 steps
  • ComputeCUDA (A100 noted)

What’s public vs private

Public now

  • - Core resilient controller implementation and experiments in the public repo
  • - Bench code and environment scaffolding

Private for now

  • - Any unreleased evaluation extensions and internal benchmarking harness
  • - Commercial product packaging and integration glue

Research areas

Resilience controllers for RL agents
Bounded-state safety and recovery
Biologically-inspired coping signals
Graph-based multi-agent control

Talk

Resilient Agent + Resilient Controller (Coping Hormones)

NolaAi · 2025-11-11

Want slides or a walkthrough? Reach out via the contact page.

Publications and technical reports will appear here.

Join the waitlist

Get early access updates for Memory Proxy, memory conversion, and SDK releases.