Tyler A. Young 🧑🏻‍💻⚗️

#ElixirLang developer working at Jump. Former software dev at @felt and X-Plane. He/him.

Public Key

npub1dwca0vsahtluny9645zl3uxcfq8k3gesh64ttxryya05xqxneqhqfdgd29

NIP-05 Address

tylerayoung@fosstodon-org.mostr.pub

Profile Code

nprofile1qqsxhvwhkgwm4l7fjza26p0c7rvysrmg5vcta244npjzwh6rqrfustsqwadd0

Author Public Key

npub1dwca0vsahtluny9645zl3uxcfq8k3gesh64ttxryya05xqxneqhqfdgd29

Show more details

Last Notes

2025-05-30 20:34:27

One of y’all should go write #ElixirLang at Apple.
https://jobs.apple.com/en-us/details/200604960/sr-software-engineer-elixir-environmental-systems?team=SFTWR
#ElixirJobs

2025-05-27 12:38:14

TIL that SmartRent forked the #ElixirLang Styler project to actually make it configurable. Quokka is the Credo-but-just-fix-it-for-me that I've wanted for 6 years of doing Elixir. 😍 😍 😍
https://github.com/smartrent/quokka

2025-05-22 20:50:03

Why won't my coworkers approve my PR??
https://cdn.fosstodon.org/media_attachments/files/114/553/454/725/599/768/original/a7bca6d832a473f6.png

2025-05-14 14:51:01

- reply

@nprofile…ewc6 Coming soon:
Go
Now
HB Go
HB Go Max
HB Go Max Now

2025-05-13 11:10:02

Proposal for a new convention for referring to #ElixirLang functions when you absolutely can't be bothered to count its arity:
MyModule.my_fun/???

2025-05-12 16:12:11

Being a professional programmer means becoming just incredibly well acquainted with your own fallibility.

2025-05-09 11:24:40

@nprofile…n5tq Hey, sorry to bug you… if you can approve this fix to CI, I can take care of the remaining open PRs on Geo. 😄
https://github.com/felt/geo/pull/235

2025-04-29 10:23:48

Yay! The FLAME talk @nprofile…aq6x gave at ElixirConf US last year is finally up. This was one of my favorites of the conference.
https://youtu.be/KrAqMyjbkJQ
#ElixirLang

2025-04-24 15:31:55

- reply

@nprofile…knxn I think it's Safari. It's not machine-specific, because every Mac I've used has been like this, with GitHub only.

2025-04-16 20:28:52

There's a remote code execution vulnerability in OTP. Update your #ElixirLang deployments to OTP 27.3.3, 26.2.5.11, or 25.3.2.20 ASAP.
https://github.com/erlang/otp/security/advisories/GHSA-37cp-fgq5-7wc2

2025-04-16 16:20:06

- reply

This is unintuitive to ~100% of people when they first learn about it, but it sort of makes sense, in that you’d like to know your changes will be okay after you merge, not just on your branch.

2025-04-16 16:20:05

PSA: GitHub Actions always runs CI against a temporary merge of your branch + your main branch. The surprising side effect of this is that tests that don't even exist on your branch can fail in CI (because you’re behind `main`).

2025-04-14 16:03:01

#ElixirLang people, has anyone actually measured the cost of larger stacktraces in production? It seems to cost approximately nothing, but it makes a huge difference in debugging. I'd love to see the Phoenix default of 20 apply to all environments, not just dev.

2025-04-14 13:52:12

64 GB of RAM. 16 CPU cores. Can't type in a comment box on a GitHub PR review in real time.
Maybe someday we'll have the computing resources for this...

2025-04-13 11:29:37

My 8 year old has been really into making games in Scratch, and is thinking about moving to a text-based programming language to get past some of Scratch's limitations.
Any recommendations for simple Elixir-based games, or should I bite the bullet and do Python?
#ElixirLang

2025-04-12 11:21:23

I have an unhealthy love of utterly unwieldy conference talk titles. They just amuse the hell out of me.
Current title I'm playing around with:
You Already Know What a Monad Is (Even If You Didn’t Know That You Know), and Why That Matters

2025-04-03 18:06:30

I missed it initially, but the call for proposals is out for ElixirConf US 2025.
I think I'm gonna submit a couple talks, one a followup to my lightning talk last year about the Result monad, and another about adversarial web scraping.
https://sessionize.com/elixirconf-2025/

2025-04-02 15:20:22

- reply

I've documented what I ended up doing here:
https://github.com/elixir-lang/elixir/pull/14385#issuecomment-2772915205

2025-04-01 13:10:12

- reply

Likewise I tried adding a blocklist derivation of Inspect to `MyApp.Schema`:
defmodule MyApp.Schema do
defmacro __using__(opts) do
quote do
use Ecto.Schema
@derive {Inspect, except: [:password, :name]}
end
end
end
...but Inspect gets mad about unknown fields.
Any ideas?

2025-04-01 13:09:44

Okay #ElixirLang folks, I need your ideas...
I'd like to override the fallback Inspect protocol for all structs to redact certain sensitive keys. For instance, I'd like the backtrace of:
Enum.reduce(%User{password: "secret"}, fn _, _ -> :ok end)
...to redact "secret".
I'd really like to avoid relying on devs to remember to derive the Inspect protocol on every struct they create.
I tried to do:
defimpl Inspect, for: Map do
...
end
...but Elixir is displeased with me (inscrutable errors).

2025-03-31 14:02:00

Hey, uh... apologies to everybody I've been less than responsive to, especially around open source stuff. My dad died unexpectedly at the beginning of March, and literally 8 hours later, my grandma went to the ICU for three brain bleeds. Family stuff has been A Lot™. 😬

2025-03-20 13:42:20

Man, I feel this so much. I've invested a ton of time into GitHub Actions over the years, and the tight integration with GitHub makes it really hard to beat (imagine that). There are so many sharp corners, though, on seemingly simple things.
https://www.feldera.com/blog/the-pain-that-is-github-actions

2025-03-18 14:58:32

- reply

Obviously this isn’t a huge optimization, but it’s nice to know about, and it can make a speed difference on a big enough test suite.
The biggest limitation is that it won’t run checks that only happen within the actual database, like foreign key checks.

2025-03-18 14:58:32

- reply

Example:
test "changeset accepts default action" do
assert %MyModule{action: :default} =
MyModule.changeset(%{action: :default})
|> Ecto.Changeset.apply_action!(:insert)
end

2025-03-18 14:58:32

If you're writing tests for an Ecto schema module, consider using Ecto.Changeset.apply_action/2 (or its ! variant) rather than actually hitting the database.
It’ll return the struct as though it had hit the DB if and only if the changeset was valid.

2025-03-04 14:31:26

When Saša Jurić speaks, I listen.
https://www.youtube.com/watch?v=AYUNI2Pm6_w

2025-03-04 14:27:46

- reply

@nprofile…5n4p I'd recognize @nprofile…8kqq anywhere! 😁

2025-02-26 15:30:50

- reply

@nprofile…dppm I do tell Siri she's utterly useless. 😆

2025-02-26 14:47:35

I am impeccably polite when addressing an LLM. Some day, when our AI overlords take over, they're going to remember that and spare me.

2025-02-12 17:09:36

Me writing TypeScript:
https://cdn.fosstodon.org/media_attachments/files/113/992/019/102/629/511/original/f7cc3ee361210150.jpg

2025-02-11 16:01:10

We got headshots at work... should I switch my profile pic?
https://cdn.fosstodon.org/media_attachments/files/113/986/087/789/357/571/original/7a6df10cfb1436e8.jpg