| package | what it does | stars |
|---|---|---|
| forgeplan | HTN + MCTS planning engine for LLM agents — separates planning from generation | |
| infiltr | RL-based adversarial red-teaming for LLMs with MITRE ATLAS reporting | |
| alphasig | Causal signal extraction from SEC EDGAR filings — not correlation, actual causation | |
| prediq | Cross-exchange prediction market analytics and arbitrage detection | |
| jaxphys | Differentiable JAX physics engine — classical, quantum, EM, fluid in one API | |
| floww | Chaos engineering toolkit for AI agent pipelines — know your reliability before prod | |
| a2a-mesh | Multi-agent mesh runtime — A2A protocol, semantic routing, DAG workflows | |
| agentomaly | Runtime behavioral anomaly detection for AI agents via OpenTelemetry | |
| leancode | Formally verified code generation — Lean 4 proof obligations, zero silent bugs | |
| healthpipe | HIPAA-compliant clinical data pipeline with differential privacy + synthetic data |
Full-stack Developer & AI Engineer | Building distributed agent meshes, JAX-based physics engines, and semantic search utilities
-
Student
- Jacksonville, FL
- https://github.com/sushaan-k
Highlights
- Pro
Pinned Loading
-
agentomaly
agentomaly PublicBehavioral anomaly detection and runtime observability for AI agents
Python 4
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.




