Alex

Software Engineer at @enosys_global

[Alex @no_niked]

Only piece that I didn't figure out is how it would replace me in sending memes in response to PRs

[Alex @no_niked]

Promise myself to not be lazy that time, and replace myself with AI since Monday

[Alex @no_niked]

My pain is that all computer use tools so far fail miserably once they need to use Chrome extension.

[Alex @no_niked]

@samwillis But who's going to be reviewing it ?

[Alex @no_niked]

Composer 2.5 is like savior for my ADHD habbits. When I miss something in my mental model for my code, it's so fast that you basically have no delay on research tasks and it's beautiful

[Alex @no_niked]

wow, cursor has become much better, I'm really getting used to Composer 2.5 for more in-depth reviews. Wouldn't use it for broader design / architectrure tasks, but purely for code it rocks. Great job @cursor_ai

[Alex @no_niked]

@theo Rick and morty flashbacks

[Alex @no_niked]

... arguments on desktop apps, and I'm willing to try

[Alex @no_niked]

Comparing Codex Desktop/T3 Code and Cursor that week. I've used agents in cli mostly and at some point it brought me so much peace to have almost everything like agents and editors / tools grouped by workspaces in my terminal. But I hear people's ...

[Alex @no_niked]

wow, cursor has become much better, I'm really getting used to Composer 2.5 for more in-depth reviews. Wouldn't use it for broader design / architectrure tasks, but purely for code it rocks. Great job @cursor_ai

[Alex @no_niked]

yeey programming is solved

[Alex @no_niked]

... chunks of tasks and manages to plan 10 PRs for rather small feature.

[Alex @no_niked]

I have so mixed feelings about subagents in claude code. If normally claude would cut a lot of corners and will try to ship a working demo as fast as possible, then using subagents (and superpower's subagents) it splits work in so much meaningless not-connected ...

[Alex @no_niked]

Funnily hype of html instead of md output started 2 days after I started to refactor all my skills to make my understanding of LLM output easier lol, but I really thought that it would be too bold to just ask it to write raw html lol

[Alex @no_niked]

@zhenyaThinks It should go under Siri folder, so it will be easier to throw away everything alltogether

[Alex @no_niked]

I've just dicovered cmux cli, and it's so much better already then other agent's shells and even subagents. I'm currently evaluating that but omg why didn't I see that sooner

[Alex @no_niked]

and after few times of that loop now you're running 2 research teams and developing 3 new agents before breakfast because you didn't want to copy 2 jsons manually

[Alex @no_niked]

You know that Sunday morning, where you want to do something real quick, but don't like that you have to do something manually, open claude to make a small tool for that, then you see something that you can do better in claude setup, you press that cmd+d and ...

[Alex @no_niked]

Also it's so slow up to a point where I just have to skip some steps I usually include in workflow, because I know that it will be taking 15 minutes to read something from that webpage

[Alex @no_niked]

I might not get something, but Claude Cowork gives me much worse experience then any harness overall with the same Opus 4.7. Like man, it just feels soo much dumber like I talk to 1 year old model.

[Alex @no_niked]

I just don't get how people get excited about code and used desktop app for generally agentic tasks, it's always huge downgrade for what it can actually do