Getting Started with Cline—The Best VS Code AI Plugin!

Guys I’ve been cheating…

On Neovim.

It makes me sad but the truth is that open-source AI tools like Cline are so much farther ahead of the Neovim AI options.

If you haven’t tried Cline yet then you gotta check it out—it’s ‘Cursor on steroids’, as my buddy described it.

Check out the video on YouTube

And hey, at least VS Code has Vim key bindings!

AI Engineer Roadmap Launch 🚀

The AI Engineer Roadmap is live! You can sign up and start tracking your progress on https://zazencodes.com/

I’ve got a discount code that’s valid during launch month—use ZAZEN33 to get 33% off (ends March 31st 2025)

Topics from this week’s video

• Introduction to Cline

  • Autonomous coding agent plugin for VS Code

  • Compared to Cursor in functionality

  • Testing its performance with a real-world use case

• Use Case: Dockerizing a RAG System

  • Application uses Google’s Vertex AI for retrieval-augmented generation (RAG)

  • Goal: Deploy the application to Google Cloud using Docker

  • Steps involved:

    • Adding a Dockerfile and Docker Compose

    • Exposing the application via an API using FastAPI

    • Updating the README with relevant instructions

• Setting Up Cline

  • Installed Cline from the VS Code Extensions Marketplace

  • Selected model for execution:

    • Started with Ollama running Gemma 2

    • Switched to Claude after performance issues

• First Attempt with Local Models

  • Started with Ollama and a 9B parameter model

  • Observations:

    • High GPU usage (100%)

    • Prompt engineering details in HTML formatting

    • Failed execution due to model limitations

  • Tried different models:

    • QN 0.5B (failed)

    • QN 14B (slow and ineffective)

    • Considered QN 32B, but impractical for local use

• Switching to Claude

  • Chose Claude 3.7 via Anthropic API

  • Significant improvements:

    • Faster response time

    • Increased context window

    • Real-time tracking of API costs per request

  • Execution steps:

    • Generated Docker configuration files (Dockerfile, Docker Compose)

    • Created and updated application files

    • Added FastAPI endpoint

    • Wrote test cases

• Evaluating Cline’s Output

  • Reviewed generated files:

    • Dockerfile with proper environment setup

    • Docker Compose configuration with correct ports and environment variables

    • README file updated with API usage instructions

  • Tested API:

    • Successfully ran Docker container

    • Performed API health check

    • Executed sample queries via curl

    • Debugged credential issues

• Final Thoughts on Cline

  • Pros:

    • Highly capable, especially with Claude

    • Well-structured prompt engineering

    • Real-time cost tracking and efficient logging

    • Open-source and Apache licensed

  • Cons:

    • Expensive when using Claude (~$0.30 for this task)

    • Local models underperform or are impractical

    • Best suited for large-scale, long-running tasks

---

Everything is fate in our lives, it's whether or not you can take that step forward when fate presents itself.

人生は巡り合わせです。巡り合わせの時に一歩踏み出せるかどうか。

Victor, Stamina Ichiban, Osaka