Mar 19
Loading video player...
AI coding agents are powerful but fundamentally limited by their training data — they have no awareness of your company's internal standards, custom abstractions, or architecture decisions. This video explores that gap and walks through progressively better solutions, from unreliable tribal knowledge to copy-pasting docs, and finally to the real answer: a RAG (Retrieval-Augmented Generation) pipeline that gives AI access to your entire organizational knowledge base. Using Kubernetes deployments as a practical example, the video demonstrates how documents are chunked, converted into embeddings, and stored in a vector database so that AI can retrieve relevant internal context automatically when answering questions or generating code. The demo showcases DevOps AI Toolkit, which includes an MCP server for knowledge management and a Kubernetes controller that continuously ingests documents from Git repositories. The results are striking — instead of producing generic Kubernetes manifests, the AI automatically applies the organization's custom App CRD, correct namespaces, scaling defaults, and security policies without anyone having to remember or specify them. The video also covers honest trade-offs, including infrastructure complexity, chunk quality challenges, and the risk of stale data, while making the case that a well-maintained RAG pipeline is the only approach that truly scales across teams and knowledge types. ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Sponsor: Kilo Code 🔗 https://kilo.ai ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #AIcodingagents #RAGpipeline #Kubernetes Consider joining the channel: https://www.youtube.com/c/devopstoolkit/join ▬▬▬▬▬▬ 🔗 Additional Info 🔗 ▬▬▬▬▬▬ ➡ Transcript and commands: https://devopstoolkit.live/ai/ai-coding-agents-are-blind-to-your-company-knowledge-heres-the-fix 🔗 DevOps AI Toolkit: https://devopstoolkit.ai ▬▬▬▬▬▬ 💰 Sponsorships 💰 ▬▬▬▬▬▬ If you are interested in sponsoring this channel, please visit https://devopstoolkit.live/sponsor for more information. Alternatively, feel free to contact me over Twitter or LinkedIn (see below). ▬▬▬▬▬▬ 👋 Contact me 👋 ▬▬▬▬▬▬ ➡ BlueSky: https://vfarcic.bsky.social ➡ LinkedIn: https://www.linkedin.com/in/viktorfarcic/ ▬▬▬▬▬▬ 🚀 Other Channels 🚀 ▬▬▬▬▬▬ 🎤 Podcast: https://www.devopsparadox.com/ 💬 Live streams: https://www.youtube.com/c/DevOpsParadox ▬▬▬▬▬▬ ⏱ Timecodes ⏱ ▬▬▬▬▬▬ 00:00 Internal Knowledge for AI 01:18 Kilo Code (sponsor) 02:48 AI Without Company Context 06:12 RAG Pipeline Explained 08:43 Knowledge Ingestion Setup 12:52 RAG for Company Knowledge