Building Enterprise-ready CI/CD Pipelines using Agentic AI
summary
Claims enterprise pipelines must be scalable secure and compliant with 80% of SDLC failures happening after code generation not during it; proposes a hybrid RAG + knowledge graph as compiler built from extracted entities metadata and relationships; shows a working Harness implementation where a Harness unified agent splits work into specialized agents, enforces arbback and policy guardrails, maintains memory and audit trails, and outputs PRs pipelines and dashboards that pass governance checks.
AI-written summary · about
chapters (10)
- 0:00Introduction to Enterprise Ready CICD Pipelines
- 0:34Understanding Enterprise Readiness
- 3:05Defining an Enterprise Pipeline
- 4:43Role of Agentic AI in Pipelines
- 5:43Context Engineering vs Prompt Engineering
- 7:01Building Knowledge Graphs
- 8:56Governance in AI-Driven Pipelines
- 10:54Enterprise Adoption Framework
- 11:34Harness Platform Overview
- 13:23Conclusion and Final Thoughts
related talks
29mOpen-Source Framework for AI-Powered CI/CD Pipelines
30mIntelligent Pipelines: Bringing AI/ML into DevSecOps Automation
44mAgentic Delivery, Guardrailed: AI in CI/CD & Platform Engineering (Open Source Patterns)
21mAI-driven optimization of CI/CD pipelines
24mNext Gen Continuous Delivery Updates
36mGenerative DevOps
34mBeyond CodeGen: How AI & Machine Learning Are Transforming DevOps
19mRethinking CI & CD for Agentic Workflows