Two kinds of AI users are emerging

Posted by martinald 5 days ago

Two kinds of AI users are emerging(martinalderson.com)

354 points | 339 commentspage 5

doom2 5 days ago|

I guess this is as good a thread as any to ask what the current meta is for agentic programming (in my case, as applied to data engineering). There are all these posts that make it to the front page talking about productivity gains but very few of them actually detail the setup that's working for the author, just which model is best.

I guess it's like asking for people's vim configs, but hey, there are at least a few popular posts mainly around git/vim/terminal configs.

swordsith 5 days ago||

In my opinion no frontier model is the best at everything, especially if you're having to catch it up with pre-existing information about your project or an esoteric scripting language, that being said, with Cursor you can try out all of the popular available models and get a feel for which do better with which tasks, in my experience - Codex is a okay model but use light thinking if you value your time, Gemini 3 flash is where its been at for me recently if I need to do big changes I go to that, And cursors model composer is good for making plans or doing refactors / making rules. Cursor gives you tools to make prompting feel like less of a repetition game, so you worry more about the task at hand and its been super efficient for me. I don't use proper version control so the fact it saves a history of every files dif's, and you can jump back easily in chats and regress the code base is the game changer.

fragmede 5 days ago|||

There more stuff in mine, but at the top of my ~/.claude/CLAUDE.md file, I have:

    ## Important Instructions
    
    - update todo.md as items are completed
    
    **Commit to git after making code changes.** Check `git status` first - only commit if there are actual changes:
    ```bash
    # If not in a git repository, initialize it first:
    git init
    
    # Then commit changes:
    git add <FILES_UPDATED>
    # Be surgical - add only the changes you just made.
    git commit -m "Description of changes"

This lets me have bite-sized git commits that I can marshall later, rather than having to wrangl git myself.

energy123 5 days ago||

I push most work into chat interface (attach full codebase as a single file, paste in specs, describe what I want), then copy the tasklist from chat into codex. This is to reduce codex token usage to avoid breaching weekly limits. I'd use a more agent-heavy process if I didn't care about cost.

anonymousDan 5 days ago||

'some sort of security' - oh great, security as an afterthought.

hereme888 5 days ago||

I'm still trying to wrap my head over the past decade: useful AI, self operating vehicles, real AI robots, immersive VR, catching reusable rockets with chopsticks, and of course the flying cars.

What will be the expected work output for the average future worker?

drsalt 5 days ago||

what is the source data? the author says they've seen "far more non-technical people than I'd expect using Claude Code in terminal" so like, 3 people? who are these people?

athrowaway3z 5 days ago||

> sandboxing agents is difficult

I use this amazingly niche and hipster approach of giving the agent its own account, which through inconceivably highly complex arcane tweaking and configurations can lock down what they can and cant do.

---

Can somebody for the love of god tell me why articles keep bringing up why this is so difficult?

NitpickLawyer 5 days ago||

I have antigravity in its own account and that has worked pretty well so far. I also use devcontainers for the cli agents and that has also worked out well. It's one click away in my normal dev flow (I was using this anyway before for python projects).

chrisjj 5 days ago|||

Why? Because the purported benefit is to make everything easier and faster. Not safe.

fragmede 5 days ago|||

It's a bunch of work, that takes a bunch of time, and I want it nowwwww-owwwww!

...is how I imagine that conversation goes.

DavidPiper 5 days ago||

> To really underline this, Microsoft itself is rolling out Claude Code to internal teams, despite (obviously) having access to Copilot at near zero cost, and significant ownership of OpenAI. I think this sums up quite how far behind they are

I think it sums up how thoroughly they've been disrupted, at least for coding AIs (independent of like-for-like quality concerns rightly mentioned elsewhere in this thread re: Excel/Python).

I understand ChatGPT can do like a million other things, but so can Claude. Microsoft deliberately using competitors internally is the thing that their customers should pay attention to. Time to transform "Nobody gets fired for buying Microsoft" into "Nobody gets fired for buying what Microsoft buy", for those inclined.

nickphx 5 days ago||

Three kinds, those who do not use it.

anxoo 4 days ago|

yes, the third kind of AI user. kind of similar to the three kinds of soccer players: offence, defence, and people who don't play soccer. thanks for the comment!

riazrizvi 4 days ago||

I think the main divisions are 1) people who are using it in fields they have little knowledge of to get basic competence, 2) the same people using it for advanced competence who are kidding themselves, 3) experts who are battling it in their own fields to finally get better answers than they could without it.

The first group are like Improved-Generalists. The third are Improved-Specialists. The second are delusional hype jockeys that drive the dumb talking points that extrapolate up the whazoo what AI is going to do and whatnot.

okokwhatever 5 days ago|

You can see the fear all around this thread. And, tbh, it makes total sense. There is nothing we can do to stop this dropping ball, we can accept it or leave the room but the industry has changed for all of us. I mean, you can use it one way or another but the concept of critical thinking is our only survival tool if your relaying on a it job this days. How long it will last? Who cares, we're fucked anyways...

enemyz0r 4 days ago|

Yep

More comments...