iiinova — where your digital self keeps evolving

Alex Martsinovich @distantprovince.bsky.social · 29/09/2026

4. Team Fallout Agents weaken the social fabric of engineering teams by reducing interpersonal communication, increasing self-reliance, and hindering skill expression. We're all too busy prompting Claude. And prompting is an intimate process.

121

Alex Martsinovich @distantprovince.bsky.social · 29/09/2026

3. Deskilling AI erodes existing skills, impedes learning, and doesn't offer any meaningful skill progression. We can talk all we want about being architects and high-level thinkers, but the way we acquired those skills is going away, and it's unclear if there's a new one.

110

Alex Martsinovich @distantprovince.bsky.social · 29/09/2026

2. Alienation Engineers become alienated from code and, as a result, care less. Meat proxy, Claude condom – there's no shortage of colourful analogies. AI widens the distance between code and its creator, inevitably leading to a declining sense of ownership.

110

Alex Martsinovich @distantprovince.bsky.social · 29/09/2026

1. Slop LLM-generated code has a strong smell that makes codebases repulsive to humans. Bigger and better models don't seem to help. The big bet here is that we'll never have to read code again, but so far, that hasn't happened, and we're having a very bad time.

110

Alex Martsinovich @distantprovince.bsky.social · 29/09/2026

Agentic coding is undeniably very useful. It's also very bad. I compiled a list of what I think is wrong with it: four problems with no solution in sight. Or, as I call them, the four horsemen of agentic coding. ⬇️

130

Alex Martsinovich @distantprovince.bsky.social · 28/09/2026

please ask people how they feel as well

000

Alex Martsinovich @distantprovince.bsky.social · 26/09/2026

alright new post goes live next week. it'll be about everything I think is wrong with agentic coding

120

Alex Martsinovich @distantprovince.bsky.social · 25/09/2026

but what will you say when one day you don't remember how to exit vim

120

Alex Martsinovich @distantprovince.bsky.social · 18/09/2026

at this point i'm more interested in what _isn't_ a tool. because everything is kinda a tool

130

Alex Martsinovich @distantprovince.bsky.social · 11/09/2026

you know what they say about the second best time

020

Alex Martsinovich @distantprovince.bsky.social · 09/09/2026

Forgot to mention but Chatting Alone got an honorable mention in Adam Mastroianni's annual blog competition! Adam is brilliant and his own writing is outrageously good, so you can imagine how huge this is for me ❤️

100

Alex Martsinovich @distantprovince.bsky.social · 05/09/2026

I'm so confused, the landing page says it's built for humans and is going to block slop, but a lot of the copy is full of claudisms? Especially faq section

060

Alex Martsinovich @distantprovince.bsky.social · 04/09/2026

haha yeah they can be a nice touch. although i think AI writing isn't exactly subtle, so everybody who cares learn to smell it very quickly

000

Alex Martsinovich @distantprovince.bsky.social · 03/09/2026

Well yes that would be for normal prose, but LLM generated talks is an underdeveloped genre, so I don't know what I'm working with here exactly! I apologize for inconvenience

010

Alex Martsinovich @distantprovince.bsky.social · 03/09/2026

well I unfortunately will have to watch it before I know how upset I should be, but sounds concerning!

100

Alex Martsinovich @distantprovince.bsky.social · 03/09/2026

let llm write your next blog post, that might do it!

110

Alex Martsinovich @distantprovince.bsky.social · 02/09/2026

Elixir is on the HN front page, but for a sad reason news.ycombinator.com/item?id=4953...

news.ycombinator.com

000

Alex Martsinovich @distantprovince.bsky.social · 01/09/2026

Some of us believe that explanations are more like puzzle pieces that need to fit juuuust right into people's minds!

010

Alex Martsinovich @distantprovince.bsky.social · 28/08/2026

We should just add new Unicode characters for Claudisms

000

Alex Martsinovich @distantprovince.bsky.social · 28/08/2026

Thanks! Last question if I may... How do you feel? It sounds like a ton of review work and a very common complaint is that reading LLM prose and code all day every day can be pretty hard.

100

Alex Martsinovich @distantprovince.bsky.social · 28/08/2026

I see. Did it have any effect on the artifact that comes before the plan, e.g. issue tracking ticket? I would imagine the initial problem should be fairly underdeveloped, otherwise it would overlap with the plan quite a bit?

100

Alex Martsinovich @distantprovince.bsky.social · 28/08/2026

Thanks you for the write up! not a lot of real team reports out there. Am I getting this right, you replaced single PR review with two (for the plan and the implementation)?

100

Alex Martsinovich @distantprovince.bsky.social · 27/08/2026

You are a better person than me, I would be filled with resentment by the end of week 1 😄

020

Alex Martsinovich @distantprovince.bsky.social · 27/08/2026

Thanks for the update! Looking forward to more, I'm still super curious about your experience being in a team of AI-coders. Seems like so far this wasn't a problem!

120

Alex Martsinovich @distantprovince.bsky.social · 24/08/2026

oh they are amazing at improving human-written code, that's for sure!

010

Alex Martsinovich @distantprovince.bsky.social · 21/08/2026

We really need to tone down the "developers feel their identity threatened" argument. This is very condescending and borderline bad faith. LLMs have a laundry list of negative effects, we can talk about them for years and won't get even close to identity issues.

010

Alex Martsinovich @distantprovince.bsky.social · 18/08/2026

Among others, I think there's also an input/output spectrum that people fall onto. You don't want to be at either extreme.

000

Alex Martsinovich @distantprovince.bsky.social · 16/08/2026

LLMs might bring a ton of bad ideas to life, sure, but what if they also leave a lot of good ideas starved of effort and ultimately underdeveloped

022

Alex Martsinovich @distantprovince.bsky.social · 13/08/2026

Why do you think people use it? You listed quite a few drawbacks!

100

Alex Martsinovich @distantprovince.bsky.social · 13/08/2026

does it work for you in a team setting? I feel like being a tradcoder in a team of ai-assisted developers is a recipe for resentment

100

Alex Martsinovich @distantprovince.bsky.social · 13/08/2026

If AI models really _are_ characters/celebrities, there's going to be a constant demand for new ones even if capabilities remain the same. Infinite reality show for tech folks

000

Alex Martsinovich @distantprovince.bsky.social · 13/08/2026

Oh I wrote about this too! Curious what's the context?

100

Alex Martsinovich @distantprovince.bsky.social · 03/08/2026

nooo not the dog wearing pants situation

020

Alex Martsinovich @distantprovince.bsky.social · 03/08/2026

that's probably #1 thing that drives me crazy. i suspect they were tortured during RL if their programs unexpectedly terminated, so they just can't comprehend a let it crash way

020

Alex Martsinovich @distantprovince.bsky.social · 03/08/2026

There's this feeling in the air that something has to be done about LLMs but it isn't clear what and on what grounds. Today I explore the frame that LLMs extract value from our shared past and pollute the future. And the past is a natural resource. distantprovince.substack.com/p/the-past-i...

distantprovince.substack.com

The Past Is a Natural Resource

And I fear it might not be renewable

040

Alex Martsinovich @distantprovince.bsky.social · 02/08/2026

New substack post tomorrow, yay! Writing is a horrible activity, absolute torture. I see why people happily outsource to LLMs.

000

Alex Martsinovich @distantprovince.bsky.social · 01/08/2026

yeah of course, but how do we ensure that humans can use LLMs responsibly if they never get to make mistakes? or rather, if they never get to make anything themselves at all? pre-AI engineers are in limited supply!

130

Alex Martsinovich @distantprovince.bsky.social · 26/07/2026

Last Code Beam Lite I've been to barely had any AI related talks, and it felt really bizarre. Like reading an old newspaper, something from the past.

110

Alex Martsinovich @distantprovince.bsky.social · 17/07/2026

Thanks for this post! Have you already seen the impact of 2x productivity boost on business metrics?

100