Maciej Gryka
@maciej.gryka.net
Software generalist with a particular interest in ML and AI. Soft spot for #ElixirLang and #Python. I enjoy challenges of early-stage startups. Ex-CTO, founded a couple companies, PhD in computer vision.
More at https://maciej.gryka.net/
More at https://maciej.gryka.net/
we just released small-and-mighty distilled PII reduction models, check them out github.com/distil-labs/...
you can run them locally (1B!) and they work as well as much bigger generic models; SLMs FTW!
you can run them locally (1B!) and they work as well as much bigger generic models; SLMs FTW!
GitHub - distil-labs/Distil-PII
Contribute to distil-labs/Distil-PII development by creating an account on GitHub.
github.com
October 16, 2025 at 3:29 PM
we just released small-and-mighty distilled PII reduction models, check them out github.com/distil-labs/...
you can run them locally (1B!) and they work as well as much bigger generic models; SLMs FTW!
you can run them locally (1B!) and they work as well as much bigger generic models; SLMs FTW!
I'll give a demo at the Berlin Elixir Meetup tonight! www.meetup.com/elixir-berli... (last minute change)
Come say hi if Elixir and/or AI is your jam. I'll show how to use distillation (featuring distil labs ofc) to make exmeralda.chat run on a smaller model.
#myelixirstatus
Come say hi if Elixir and/or AI is your jam. I'll show how to use distillation (featuring distil labs ofc) to make exmeralda.chat run on a smaller model.
#myelixirstatus
October 9, 2025 at 10:26 AM
I'll give a demo at the Berlin Elixir Meetup tonight! www.meetup.com/elixir-berli... (last minute change)
Come say hi if Elixir and/or AI is your jam. I'll show how to use distillation (featuring distil labs ofc) to make exmeralda.chat run on a smaller model.
#myelixirstatus
Come say hi if Elixir and/or AI is your jam. I'll show how to use distillation (featuring distil labs ofc) to make exmeralda.chat run on a smaller model.
#myelixirstatus
Over the last couple of weeks we've been working on adding tool calling support to our distillation pipelines. Now you can use it!
As a fun demo, we trained a model to help you remember git commands and it's only 3B so you can run it locally!
www.distillabs.ai/blog/gitara-...
As a fun demo, we trained a model to help you remember git commands and it's only 3B so you can run it locally!
www.distillabs.ai/blog/gitara-...
Gitara: How we trained a 3B Function-Calling Git Agent for Local Use
How we created a local tool-calling language model to turn plain-English language questions into git commands with the accuracy of a cloud LLM. You can check it out in our GitHub repo or get the model...
www.distillabs.ai
October 1, 2025 at 3:07 PM
Over the last couple of weeks we've been working on adding tool calling support to our distillation pipelines. Now you can use it!
As a fun demo, we trained a model to help you remember git commands and it's only 3B so you can run it locally!
www.distillabs.ai/blog/gitara-...
As a fun demo, we trained a model to help you remember git commands and it's only 3B so you can run it locally!
www.distillabs.ai/blog/gitara-...
If you're skeptical of Big AI and want to learn more about Small AI, check out the blog post our CTO, Jacek, just published. Goes through how we can help you turn your big, expensive, small models into beautiful, small, efficient ones www.distillabs.ai/blog/small-e...
Distil Labs
www.distillabs.ai
September 16, 2025 at 6:50 AM
If you're skeptical of Big AI and want to learn more about Small AI, check out the blog post our CTO, Jacek, just published. Goes through how we can help you turn your big, expensive, small models into beautiful, small, efficient ones www.distillabs.ai/blog/small-e...
Reposted by Maciej Gryka
In September, we are hosting Elixir Berlin as part of the #globalelixirmeetups. Join us and other Elixir developers around the globe to celebrate this week!
📅 Hope to see you there! 💜
#ElixirLang
www.meetup.com/elixir-berli...
📅 Hope to see you there! 💜
#ElixirLang
www.meetup.com/elixir-berli...
Sep 25th, [In-Person] Elixir Meetup #112 - Global Elixir Meetup Edition, Thu, Sep 25, 2025, 7:00 PM | Meetup
👋 Hallo und 💜-lich Willkommen to the Elixir Berlin Meetup.
This month we are joining Global Elixir Meetups!
We will have three talks scheduled at a new first time locat
www.meetup.com
September 5, 2025 at 7:40 AM
In September, we are hosting Elixir Berlin as part of the #globalelixirmeetups. Join us and other Elixir developers around the globe to celebrate this week!
📅 Hope to see you there! 💜
#ElixirLang
www.meetup.com/elixir-berli...
📅 Hope to see you there! 💜
#ElixirLang
www.meetup.com/elixir-berli...
Reposted by Maciej Gryka
Phoenix.new by @fly.io is an interesting new entrant into the prompt-driven app development space - it can spin up a full Elixir/Phoenix real-time web app in a neat browser-based IDE, testing via a headless browser while it writes the code
simonwillison.net/2025/Jun/23/...
simonwillison.net/2025/Jun/23/...
Phoenix.new is Fly’s entry into the prompt-driven app development space
Here’s a fascinating new entrant into the AI-assisted-programming / coding-agents space by Fly.io, introduced on their blog in Phoenix.new – The Remote AI Runtime for Phoenix: describe an app in …
simonwillison.net
June 23, 2025 at 6:45 PM
Phoenix.new by @fly.io is an interesting new entrant into the prompt-driven app development space - it can spin up a full Elixir/Phoenix real-time web app in a neat browser-based IDE, testing via a headless browser while it writes the code
simonwillison.net/2025/Jun/23/...
simonwillison.net/2025/Jun/23/...
Reposted by Maciej Gryka
Oban is probably the greatest library I have ever used, especially with Oban.Pro on top. The value you get for free - or for a very reasonable amount - is mind-blowing.
June 20, 2025 at 2:24 PM
Oban is probably the greatest library I have ever used, especially with Oban.Pro on top. The value you get for free - or for a very reasonable amount - is mind-blowing.
Claude just looooooves monkey-patching all over Python tests even though the rules tell it not to do that.
June 13, 2025 at 1:00 PM
Claude just looooooves monkey-patching all over Python tests even though the rules tell it not to do that.
Has there even been a better/worse timing of two papers published the same day?
May 26th 10am: x.com/xuandongzhao... "Learning to Reason without External Rewards: LLMs can learn complex reasoning without access to ground-truth answers, simply by optimizing their own internal sense of confidence."
May 26th 10am: x.com/xuandongzhao... "Learning to Reason without External Rewards: LLMs can learn complex reasoning without access to ground-truth answers, simply by optimizing their own internal sense of confidence."
x.com
May 27, 2025 at 7:42 PM
Has there even been a better/worse timing of two papers published the same day?
May 26th 10am: x.com/xuandongzhao... "Learning to Reason without External Rewards: LLMs can learn complex reasoning without access to ground-truth answers, simply by optimizing their own internal sense of confidence."
May 26th 10am: x.com/xuandongzhao... "Learning to Reason without External Rewards: LLMs can learn complex reasoning without access to ground-truth answers, simply by optimizing their own internal sense of confidence."
Truly an excellent name. I don’t think I’ve ever encountered a better project name.
April 27, 2025 at 7:47 PM
Truly an excellent name. I don’t think I’ve ever encountered a better project name.
In case you doubted "all devs will become managers", I'm currently asking Claude via the CLI to plan and implement a bunch of changes and then Gemini via Windsurf to do code review of the changes.
My first thought was "ROBOT FIGHT", but actually they're cordial and take each other's feedback well.
My first thought was "ROBOT FIGHT", but actually they're cordial and take each other's feedback well.
April 17, 2025 at 9:48 AM
In case you doubted "all devs will become managers", I'm currently asking Claude via the CLI to plan and implement a bunch of changes and then Gemini via Windsurf to do code review of the changes.
My first thought was "ROBOT FIGHT", but actually they're cordial and take each other's feedback well.
My first thought was "ROBOT FIGHT", but actually they're cordial and take each other's feedback well.
Woah, looks like GPT 4.1 is the first model that doesn't pepper the code with gratuitous comments every 3 lines.
April 14, 2025 at 8:14 PM
Woah, looks like GPT 4.1 is the first model that doesn't pepper the code with gratuitous comments every 3 lines.
Reposted by Maciej Gryka
I am also the author of a history of concentration camps in the Soviet Union, and I concur
As the author of a history of concentration camps, I would like to emphasize that this is a very bad idea, and we should definitely not do it.
Leavitt confirmed that the WH is working on deporting US citizens to El Salvador
April 8, 2025 at 9:11 PM
I am also the author of a history of concentration camps in the Soviet Union, and I concur
Still love Elixir, but also getting back into some Python feels like putting my old, comfy slippers back on :)
April 7, 2025 at 8:37 PM
Still love Elixir, but also getting back into some Python feels like putting my old, comfy slippers back on :)
@aaronfrancis.com somebody made a billboard for you
Seems fitting the terminal company would get a billboard next to a `shell` station… well played @warp.dev team
April 3, 2025 at 9:42 PM
@aaronfrancis.com somebody made a billboard for you
Reposted by Maciej Gryka
Ohhhhh you guys are all at @alchemyconf.com. I'M SO JEALOUS.
April 1, 2025 at 3:46 PM
Ohhhhh you guys are all at @alchemyconf.com. I'M SO JEALOUS.
Still using Windsurf as my main editor, but I find myself gravitating towards the Claude CLI tool more and more, especially for more complex changes.
I'm not sure exactly why that is, but somehow it feels more reliable/trustworthy?
I'm not sure exactly why that is, but somehow it feels more reliable/trustworthy?
April 1, 2025 at 12:37 PM
Still using Windsurf as my main editor, but I find myself gravitating towards the Claude CLI tool more and more, especially for more complex changes.
I'm not sure exactly why that is, but somehow it feels more reliable/trustworthy?
I'm not sure exactly why that is, but somehow it feels more reliable/trustworthy?
Hard-won lesson! And a hard pill to swallow, because I don’t know if being systematic about/focused on this helps much or of it’s largely luck. You’ll only know for sure after years of work and guessing up front how things will turn out seems ~impossible?
Picking your market is far more important than picking any one feature, and might be more important than every feature you will ever build.
March 25, 2025 at 9:02 AM
Hard-won lesson! And a hard pill to swallow, because I don’t know if being systematic about/focused on this helps much or of it’s largely luck. You’ll only know for sure after years of work and guessing up front how things will turn out seems ~impossible?
Reposted by Maciej Gryka
What this all really points towards, to me, is how fast code review is going to become a bottleneck. But it's not just code review, it's also implementation plan review and poking around at various constraints in a system. Claude can't produce more working software without this stuff evolving too
Whelp, filed 4 bug reports already 😅
Claude Code needs guidance from a senior domain expert to work well IMO:
- Accidentally proposed an n^2 implementation
- Didn't realize it was correct about something else
- Made a very bad assumption I had to correct
- Corrective plan doubled memory use
Claude Code needs guidance from a senior domain expert to work well IMO:
- Accidentally proposed an n^2 implementation
- Didn't realize it was correct about something else
- Made a very bad assumption I had to correct
- Corrective plan doubled memory use
Vibe check using Claude Code:
It's good. Implemented a real feature, tests, benchmarks, understood memory sensitivity too. Roughly how I would have written this myself.
NOTE: this has not undergone code review, likely missing some subtlety. But a great first step
github.com/honeycombio/...
It's good. Implemented a real feature, tests, benchmarks, understood memory sensitivity too. Roughly how I would have written this myself.
NOTE: this has not undergone code review, likely missing some subtlety. But a great first step
github.com/honeycombio/...
February 25, 2025 at 2:37 PM
What this all really points towards, to me, is how fast code review is going to become a bottleneck. But it's not just code review, it's also implementation plan review and poking around at various constraints in a system. Claude can't produce more working software without this stuff evolving too
oh shit claude code is so good
February 24, 2025 at 7:32 PM
oh shit claude code is so good
This is currently my favourite AI-based coding workflow. @windsurfai.bsky.social makes it pretty easy to just do the core of a thing (in this case, fix a bug) and let AI handle all the other ceremony around that (in this case, writing a test).
February 22, 2025 at 9:39 PM
This is currently my favourite AI-based coding workflow. @windsurfai.bsky.social makes it pretty easy to just do the core of a thing (in this case, fix a bug) and let AI handle all the other ceremony around that (in this case, writing a test).
Thanks for writing arathunku.com/b/2024/notes... @arathunku.com ! Came in handy today 👍
Notes on adding OpenTelemetry to an Elixir app
I wanted to get some telemetry data from a LiveView page.
arathunku.com
February 21, 2025 at 1:13 PM
Thanks for writing arathunku.com/b/2024/notes... @arathunku.com ! Came in handy today 👍
This is delightful. Now I want a pet rat to tickle.
I am a very normal person writing up a web compat ticket because this mouse tickling online certification course doesn't work in Firefox
nc3rs.org.uk/3rs-resource...
nc3rs.org.uk/3rs-resource...
Rat tickling: Certification
An online course created allowing you to become certified in the correct rat tickling technique.
nc3rs.org.uk
February 21, 2025 at 8:28 AM
This is delightful. Now I want a pet rat to tickle.
Reposted by Maciej Gryka
Gift link.
The actual story of the shackled people being deported without even being told where they were being taken — maybe to countries where they’d face torture or death — while their terrified children cried.😢
The White House posted that pic with “ASMR” tag.
www.nytimes.com/2025/02/18/w...
The actual story of the shackled people being deported without even being told where they were being taken — maybe to countries where they’d face torture or death — while their terrified children cried.😢
The White House posted that pic with “ASMR” tag.
www.nytimes.com/2025/02/18/w...
February 20, 2025 at 12:50 PM
Gift link.
The actual story of the shackled people being deported without even being told where they were being taken — maybe to countries where they’d face torture or death — while their terrified children cried.😢
The White House posted that pic with “ASMR” tag.
www.nytimes.com/2025/02/18/w...
The actual story of the shackled people being deported without even being told where they were being taken — maybe to countries where they’d face torture or death — while their terrified children cried.😢
The White House posted that pic with “ASMR” tag.
www.nytimes.com/2025/02/18/w...
The fact that Deep Research justifiably takes a long time makes for interesting UX. I didn’t expect to be so happy delegating work and then chilling, knowing that my report is being written. It feels nice to know something useful is happening!
February 16, 2025 at 1:34 PM
The fact that Deep Research justifiably takes a long time makes for interesting UX. I didn’t expect to be so happy delegating work and then chilling, knowing that my report is being written. It feels nice to know something useful is happening!