AI code writer

Buckle up, code warriors! AI’s grand adventure in the coding realm has been nothing short of a rollercoaster ride. Picture this: we started with humble code completion, like an eager puppy fetching snippets. Then, boom! We leveled up to out-of-band chat, where AI became our coding buddy, always ready for a brainstorming session. And now? Hold onto your keyboards, because we’ve hit the jackpot with full-blown code writing! It’s like having a tireless robot assistant who can churn out entire functions faster than you can say “Hello, World!” This AI-powered journey has turned coding from a solo trek into a wild party where everyone’s invited - from newbie coders to seasoned tech wizards. So, grab your favorite caffeinated beverage and join the fun - the future of coding is here, and it’s got AI written all over it!

• Completion
  • Co-pilot
• Coder
  • Cursor
  • Aider
  • Avante
• Use Cases
  • Using latest docs
  • Diff Summarization
  • Review changes between dates
  • Dream: Re-write commit history to break things into orthogonal changes
  • Should I code once we have AI?
    • Reasons to Program - Mastery vs Getting Shit Done
    • How can you use AI - Research Assistant vs Code Writer

Completion

Co-pilot

The original

Coder

Cursor

• Cursor Changelog
• Cursor Features

WOW - Just started playing with that, it was fantastic, super smooth, looking forward to seeing if Avante can catch up.

It basically made this whole change via chat/apply, with me making minimal suggestions

• Tab completion

  • includes the code before what you typed (huge improvement)
  • Finds the next line you likely want to change (great for sensing refactor)
• They forked VS.Code so could keep most of what it has.
• They use Claude, which seems to be better than co-pilot
• I used it mostly in chat/apply mode, looking forward to using it on a project where I’m writing more of the code (probably will try when I look at the transformer code in Jupyter again)
• You put in documentation tags to bring it, and it also does embeddings for it.

Aider

Aider Release Notes on GitHub

OK, so Aider tries to make every change its own commit, which is super noisy and error-prone, a few learnings:

1. Do the changes on a branch then squash up the final branch
2. You need to have unit tests, let Aider create them (I’ve got to figure out how to let it name them)

Here’s a change almost totally done with Aider

Avante

Avante on GitHub

Just starting to play with this. It’s a VIM plugin, looks solid.

What’s nice:

• It does a good job of inline changes/merging (like co-pilot propose diffs)
• It uses many best practices so I discovered render and other nice libraries
• Fun to see what the coding prompts are

Use Cases

Not sure if this should be a separate post, but I’m going to start looking at this from the perspective of use cases.

Using latest docs

Cursor has native support for indexing docs, some projects even have llm.txt to teach the LLM what it needs to know. Fasthtml is an example

Diff Summarization

Review changes between dates

Dream: Re-write commit history to break things into orthogonal changes

Should I code once we have AI?

It occurs to me this applies to all vocations, not just programming

Reasons to Program - Mastery vs Getting Shit Done

1. For the Love of Coding: Mastery and Joy

   • Programming can be intrinsically rewarding. The process of solving problems, creating something from nothing, and continuously improving one’s skills offers satisfaction and joy. Mastery of coding, like any craft, provides a sense of accomplishment.
   • Using VIM key combos is especially fun, like doing joystick combos in Street Fighter II - it’s a fun challenge that keeps you engaged and sharpens your skills.
   • My pleasure is coding, but I’m guessing this applies simarly to cooking, woodworking, and stampcollecting

2. To Achieve Practical Goals: Shipping Products

   • On the other hand, programming is a means to an end. It’s a tool for building and shipping products, for solving real-world problems, and for getting tangible results. Efficiency and productivity are key, and AI can help streamline these processes.

How can you use AI - Research Assistant vs Code Writer

1. Getting Help: Replacing Traditional Resources

   • AI as a resource for programming assistance is becoming increasingly popular. It provides a more interactive and immediate alternative to traditional platforms like Stack Overflow, documentation, or Reddit.
   • In the past, before the internet, finding solutions to programming problems was a cumbersome process. AI represents an evolution of these resources, offering instant, personalized help that can significantly speed up the problem-solving process.

2. AI Writing Code: The Fun vs. Automation Dilemma

   • While AI can write code efficiently, this raises the question: should we let it take over tasks we enjoy? For many, coding is a hobby, a passion akin to the thrill of executing perfect joystick combos in Street Fighter II using specific nvim key combinations. The challenge and mastery involved are part of the joy; should we willingly give that up?

Mostly lifted from this guy: