DEV Community

Krishnendu Ghosh
Krishnendu Ghosh

Posted on

# How I Built SLEK AI CLI at 15 Years Old

Building software is hard. Building an AI CLI as a student with limited resources is even harder.

Over the last few months, I've been working on SLEK AI CLI, a command-line AI assistant designed to make AI more accessible from the terminal.

Why I Started

Most AI tools today focus on web interfaces. As someone who spends a lot of time in the terminal, I wanted an AI assistant that could be used directly from the command line.

My goals were simple:

  • Fast responses
  • Multiple AI model support
  • Easy installation
  • Beginner-friendly experience

Challenges I Faced

Creating an AI CLI wasn't as straightforward as I expected.

Some of the biggest challenges included:

  • Managing API requests efficiently
  • Handling authentication securely
  • Supporting multiple AI models
  • Publishing and maintaining npm packages
  • Keeping the project free for users

Every bug taught me something new.

What SLEK AI CLI Can Do

  • Chat with AI directly from the terminal
  • Support multiple models
  • Easy npm installation
  • Lightweight and fast experience

Lessons Learned

The biggest lesson wasn't technical.

It was learning how to keep building even when things break.

Many times I thought about giving up after deployment failures, API issues, and package publishing problems. But solving those problems helped me improve as a developer.

What's Next?

I'm currently working on:

  • Better model routing
  • Improved CLI experience
  • More developer-focused features
  • Future agent capabilities

I'd Love Your Feedback

If you've built a CLI tool before:

  • What helped your project grow?
  • How did you get your first users?
  • What features do developers value most in AI CLIs?

I'd love to hear your thoughts.

Top comments (0)