ingero-io/ingero

Overview

Ingero is an eBPF agent and MCP server that provides causal observability for GPU workloads. It uses eBPF to trace GPU operations and correlate them with system events, enabling root-cause analysis of performance issues.

Best for

Best for
Developers debugging GPU performance issues in Linux-based AI/ML environments

Use cases

• Debugging GPU performance bottlenecks in AI training pipelines
• Tracing GPU memory access patterns for optimization
• Correlating GPU events with CPU-side system calls for root-cause analysis

Notes

Ingero is an eBPF agent and MCP server that provides causal observability for GPU workloads. It uses eBPF to trace GPU operations and correlate them with system events, enabling root-cause analysis of performance issues.

85 stars on GitHub. Last updated 2026-06-01. Licensed Apache-2.0.

Use cases

• Debugging GPU performance bottlenecks in AI training pipelines
• Tracing GPU memory access patterns for optimization
• Correlating GPU events with CPU-side system calls for root-cause analysis

Pros

• Low-overhead tracing via eBPF without modifying application code
• Causal correlation between GPU and system events simplifies debugging
• Open-source and written in C for performance

Cons

• Limited to Linux systems with eBPF support
• Small community and early-stage project with 85 GitHub stars
• Requires understanding of eBPF and GPU internals to interpret traces

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.