Automated SRE with Agentic Workflow and eBPF

This project brings Site Reliability Engineering into a new era by combining agentic workflows with the power of eBPF. Instead of reactive firefighting, the system acts as an intelligent agent within infrastructure—observing, diagnosing, and remediating issues in real time.

Project Category

AI Engineering

Tech Stacks:

PyTorch

Kubernetes

FastAPI

TypeScript

Python

CUDA

Ray

vLLM

AWS

View Live Demo

Github Code

Automated SRE with Agentic Workflow and eBPF This project brings Site Reliability Engineering into a new era by combining agentic workflows with the power of eBPF. Instead of reactive firefighting, the system acts as an intelligent agent within infrastructure—observing, diagnosing, and remediating issues in real time. eBPF provides deep, low-overhead visibility into kernel and application behavior, while agentic workflows orchestrate automated responses, scaling from anomaly detection to self-healing actions. The result is a proactive, adaptive SRE model where reliability isn’t just monitored—it’s continuously maintained by autonomous agents embedded at the system’s core.

PUKU CLI

PUKU CLI reimagines the command line as more than just a tool—it becomes a collaborative agent. Designed to bring agentic workflows directly into the terminal, PUKU CLI empowers developers to interact with AI-native processes seamlessly. Instead of juggling scripts, configs, and external dashboards, you can orchestrate tasks, automate flows, and query intelligent agents right where you work. It’s a fusion of traditional CLI efficiency with modern AI-driven adaptability, making the terminal not just a place to execute commands, but a space to think, build, and iterate alongside an intelligent partner.