Runtime extension
Move eBPF beyond kernel-only workflows
Run, instrument, and extend systems with userspace execution paths, GPU paths, JIT/AOT support, and deployment models that fit production tooling.
Open-source infrastructure
Eunomia
Open-source eBPF infrastructure for runtime extension, GPU tracing, and AI Agents.
What Eunomia Provides
Runtime infrastructure, public resources, and AI agent systems define the work behind Eunomia.
Resources
Publish reusable tutorials, docs, blogs, and papers
Maintain executable tutorials, documentation entry points, technical writing, and research artifacts for eBPF builders.
AI Agents
Observe and Enforce AI agents at the system layer
Zero-instrumentation, framework-agnostic visibility into what agents actually do, plus runtime enforcement for what they are allowed to do.
Projects
Open-source platform projects and public resources grouped by how people use them.
Platform
Execution engines, toolchains, runtime components, and low-level extension paths.
AI Agents
Tracing, observability, and safety work for AI agent systems.
Resources
Tutorials, documentation, blog posts, and research papers for eBPF builders.
Latest writing
All postsCan an AI Agent Tune the Linux Scheduler? Inside SchedCP
SchedCP gives AI agents a controlled path from workload intent to verified schedext policies, achieving up to 1.79x performance improvement and 13x lower optimization cost in the paper's evaluation.
Profiling AI Agents with Semantic Flamegraphs
Your AI agent spent $3000 this month. Which activities consumed that budget? agentpprof applies the flamegraph paradigm to AI agent traces, mapping natural language prompts to semantic tags and aggregating them like a CPU profiler. This post explains why existing observability tools fail at budget attribution and how semantic flamegraphs restore aggregation for agent workloads.
ActPlane: Pushing Agent Harness Enforcement Down to Kernel eBPF
ActPlane is an eBPF-based policy engine that observes and enforces AI agent behavior at the OS kernel level. This post analyzes the systemic blind spots of prompt constraints, tool-layer guards, and sandboxes, and explains how ActPlane uses label propagation and temporal predicates to implement a deterministic agent harness.