Gemini Powered CI/CD Insights: An AI Agent for Real-Time Pipeline Analysis Using Natural Language Commands

In today's fast-paced world of software delivery, keeping an eye on everything is more important than ever, especially when it comes to Continuous Integration and Continuous Deployment (CI/CD) pipelines. This paper introduces a cutting-edge AIdriven system that taps into Google's Gemini Mo...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2025 6th International Conference on Inventive Research in Computing Applications (ICIRCA) s. 1407 - 1412
Hlavní autoři: Kairuppala, Shanmukhi, Tatiparti, Druti, Putta, Sivani, Kodigudla, Krushna Varshitha, Sirapu, AshaJyothi
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 25.06.2025
Témata:
Artificial intelligence
CI/CD pipelines
DevOps tools
Measurement
Monitoring
Natural language processing
Observability
Optimization
pipeline monitoring
Pipelines
Real-time systems
Servers
Software
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In today's fast-paced world of software delivery, keeping an eye on everything is more important than ever, especially when it comes to Continuous Integration and Continuous Deployment (CI/CD) pipelines. This paper introduces a cutting-edge AIdriven system that taps into Google's Gemini Model to offer realtime insights into the health and performance of a full-stack MERN (MongoDB, Express.js, React, Node.js) application's CI/CD pipeline managed by GitHub Actions. The system automatically collects and analyzes data from GitHub Actions CI/CD logs, cloud server metrics like CPU usage, memory utilization, and network I/O via the Heroku Platform API, as well as website performance data such as page load times and error rates from New Relic. All this information is stored in a MongoDB database and can be accessed through a user-friendly chat interface built on Streamlit, allowing developers and operations teams to ask natural language questions about their system's status. The Gemini Model then processes these inquiries and the relevant data to create easy-to-understand summaries, spot anomalies, and suggest possible root causes and optimizations. Early results show that this system can greatly improve observability, cut down troubleshooting time, and support data-driven decision-making throughout the software development lifecycle.
DOI:10.1109/ICIRCA65293.2025.11089679