Embedded Software Development Efficiency Improvement with Docker and Kubernetes
summary
Demonstrates packing CI tools into Docker images so developers run the exact pipeline checks locally instead of waiting for remote jobs, shows Kubernetes used as GitLab Runner across a four-node cluster with topology/anti‑affinity tuning, and reports pipeline times dropping from ~21 minutes (standard Runner) to ~8.5 minutes (k8s default) and to 6.5 minutes after load‑balancing changes. Also outlines building a Python-based AI code‑review service that parses diffs via GitLab API and uses structured prompts and keyword filtering for consistent comments.
AI-written summary · about
chapters (7)
- 0:00Embedded dev efficiency intro
- 3:44Docker and containerization
- 8:01Docker for local dev
- 10:30Kubernetes for CI/CD
- 13:01Optimizing Kubernetes load balancing
- 17:16AI-powered code review tool
- 21:36Q&A
related talks
32mEnhance Your Kubernetes CI/CD Pipelines With Open Source
26mOptimizing Your CI Pipelines
27mOptimizing CI/CD Pipelines for Faster Time to Market
24mReinvent your pipeline with GitLab, Kubernetes, and Amazon EKS
35mGitLab CI and Kubernetes, Build Test and Deploy Your Projects
02h 47mPipes on Pipelines
54mFrom Code to Customer
33mUnderstanding GitLab CI