Alex Martsinovich

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. ⬇️

My thoughts on the current state of AI driven software development, why I'm no longer concerned about a runaway intelligence explosion, and what I think is a far worse outcome we could be walking into maevyn.codes/decoherence

Apparently, there are roughly 2 types of people: 1. "It's free, so I want all of it" 2. "It's free, so I want none of it"

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

People really hate AI written blog posts 😌😌😌 writethatblog.substack.com/p/dev-reacti...

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

We should just add new Unicode characters for Claudisms

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.

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

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

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

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...

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

There is seemingly no way to make logs look good both in dev and prod. This is especially noticeable in libraries, which often choose to simply not log. In my new post I propose a solution and explain why telemetry isn't quite it distantprovince.by/posts/what-d... #ElixirLang #MyElixirStatus

Working on (yet another one) blog post about logging and of course @hauleth.dev raised this topic on elixirforum all the way back in 2019 😆

Hello #ElixirLang Does anyone look for principal/senior level developer? Either permanent or contract. I have experience in performance optimisations and observability of Elixir and Erlang codebases. CV available at hauleth.dev/cv @sifa.id profile sifa.id/p/hauleth.dev

Proud Elixir Radar mention! 🫡

It's only thankless maintenance if it's from Nebraska region of the US, otherwise it's just sparkling resume farming

if 100 chrome tabs were a song, what a treat

It's amazing how LLM driven development is both good and bad at the same time. I don't see any extreme stance on it winning. We'll have to learn how to use this goddamn tool properly: without frying our brains or slopifying software.

I once again have some extra free time on my hands, so I'm going to dig into why DBConnectionErrors are seemingly never grouped correctly and see how many yaks require shaving. Our first stop is naive `Logger.error` calls. We need to figure out what to do with them!

📡 New Blog Post LLMs looove markdown, but the best chat interfaces go beyond and let models use rich UI elements. In this post I demonstrate one possible way to achieve this: by teaching models to use custom HTML elements. distantprovince.by/posts/teachi...

I never wrote to my MP before, but if @apenwarr.ca tells me to, I will

Wrote a bit about why LLM-written text is so painful, even when the content itself is fine, and the importance of using your own voice jola.dev/posts/the-so...

🎉 InstructorLite has passed 50k downloads on Hex! It started as a fork almost 2 years ago and quickly transformed into a full rewrite. I'm very happy that there are people (and companies!) out there who find it useful! #ElixirLang #MyElixirStatus

I dusted off my blog to explain the AI engineering policy I wrote for work. It morphed into more of a rant on the lack of leadership in tech than I intended. Yelling "MOAR AI!" is not leadership and your teams all feel it. brianmeeker.me/2026/05/14/h...

I wrote about model permanence - a very natural but false idea that models exist between conversation turns. distantprovince.substack.com/p/chatting-a...

Anyone is looking for Senior Elixir Developer based in the EU? My CV available at hauleth.dev/cv #elixirlang #job

Looking for an #ElixirLang dev? I'm officially on the market! 4+ years of Elixir in production. Natural affinity to testing, o11y and DX. My most used OS libraries are InstructorLite and PostHog SDK. Remote from 🇨🇦, PDT, links in bio. RTs appreciated 🙏 #MyElixirStatus

Shit like this made me go to therapy

Code reviews are definitely dead and many of us, including myself, are grieving. I don't know how things will turn out, but I felt the need to write a farewell. distantprovince.by/posts/a-fare...

Working on improving PostHog error tracking for complex OTP errors. Chained exceptions is a fairly good fit. Hopefully will ship soon!

Maybe AI has made writing code an obsolete skill, but every engineer who plays with coding agents today gained their experience by writing code. To understand the real impact of AI, we'll have to look at a world where engineers _never_ had to write code. distantprovince.by/posts/ai-fut...