I was supposed to stop at LangGraph.
That was the plan.
Instead, the rabbit hole decided otherwise.
Some of the things I unexpectedly picked up while chasing Agentic AI:
⚙️ Docker
- Multi-stage Dockerfiles
- Docker Compose
- Volumes
- Building images from my own projects instead of just pulling them
- Finally understanding why "works on my machine" isn't enough
🤖 GitHub Actions
- Automated builds and tests
- CI/CD pipelines
- Branch protection rules
- Secrets management
- Automatic Docker image publishing
- Realizing how satisfying it is when pushing code triggers everything for you
🌿 Git & Engineering Practices
- Conventional commits
- Pull requests even for solo projects
- Trunk Based Development
- Small, rapid merges instead of week-long branches
- Keeping main always deployable
🚀 Deployment
- Actually shipping things instead of leaving them in localhost prison
And somehow, while trying to learn AI, I ended up appreciating software engineering just as much.
Still left:
• LlamaIndex
• Cloud
• Terraform
Apparently the roadmap had other plans.
Curious what unexpected rabbit holes everyone else fell into while learning AI.
source & further reading
dev.to — original article
Testing AI Coding Agents Beyond Code Generation: A Real-World Benchmark
n8n AI Workflow Builder Brings Natural Language Automation to Cloud Workflows
I gave an AI agent a business and $0. It found the one thing AI can't sell.