akoculu

Mitte (mitte.ai) | Berlin | ONSITE, REMOTE | Full-Time

Mitte is an AI creative suite — the agentic OS for creative teams.

Engineering (Berlin, on-site):

  - Senior Full-Stack Engineer (Go, React): https://mitte.ai/careers/role/GZHn1_O6
  - Senior Frontend Engineer (React, TS, Redux): https://mitte.ai/careers/role/5qi8KR4D
  - Senior Backend Engineer (Elixir, Go): https://mitte.ai/careers/role/oaFnPm0A
  - Founding Engineer (Go, React): https://mitte.ai/careers/role/7VTbvOae

  - AI Educator (Remote): https://mitte.ai/careers/role/ZxfzRYO0
  - Founder's Associate: https://mitte.ai/careers/role/0dSWgqG1

Thank you.

Mitte (https://mitte.ai) | Berlin | ONSITE | Full-Time

Mitte is an AI creative suite. We've built our own media agent and we're hiring engineers who want this to be the best work of their career.

Founded by an ex-Superhuman lead engineer. Lean startup and unix philosophy run deep at Mitte. We ship every day and like to design systems, be thoughtful about patterns and primitives before scaling the codebase, especially with AI.

We're hiring:

- Backend Engineer (Elixir, Go) — own services end-to-end, from agent orchestration to the systems that move media at scale.

- Full-Stack Engineer (Go, TypeScript, Redux Toolkit) — build the surfaces creators actually use. Strong design taste and obsession with detail expected.

Berlin, on-site. We work in the same room because the product is moving too fast for anything else.

Reach out: careers@mitte.ai — tell us something you've built that you're proud of, and why it was hard. Berlin based applications only.

this is incredible.

We had severe issues with Bunny and recently migrated off it.

Some of our users were unable to reach our CDN altogether. They couldn't load any assets at all. Bunny's customer service was far too slow to respond and mostly gave unhelpful answers. They couldn't even identify the issue.

In less than 45 minutes, I moved our CDN entirely from Bunny to Cloudflare Workers. Now our CDN just actually works, I don't have to debug our CDN for the Bunny customer service team.

Also, this is obviously a marketing post.

cool idea, would be awesome to see a demo link

congrats!

congrats! I like the design

nice work + cool name, congrats!

I can't wait to buy my first hard copy IP-address book and look up numbers of friends & businesses in yellow pages with small letters

[flagged]

Many people lean more on reason and logic.

Yes, when theres's no impulse strong enough to outweigh reasoning. You don't need Ghazali for this, Kant also explains it. Before suggesting that I rephrase things, I think you should explore the domain first.

[flagged]

so happy to hear this! I hope you enjoy the reading. Also, feel free to check out the 4 hours long audio lecture on the book. It's really good!

And surely you read Ghazali and know what you’re talking about.

When was the last time you were threatened (“we’ll will bang on your door and take down your accounts”) by a company like Kik and you defended the principles and values you believed until the end?

I’m quite familiar with both western and eastern traditions, don’t know any better source than Ghazali’s “the alchemy of happiness” about how people make decisions.

It’s kind of ironic to see people being triggered by just the mention of it though — just reflects what your heart tends to reject impulsively. I wish you curiosity.

P.S Here’s the full quote, it entertainingly describes the ignorance here:

“first, they didn't look at the dates of the emails. They don't understand the timeline.

second, they can't relate to standing your ground in a high pressure situation involving threats.

and third, they haven't read Al-Ghazali yet, don't quite understand how (free) people make decisions”

Let me explain myself on this;

When this happened, people started making assumptions. A few of them:

1. I’m irrational 2. I’m angry 3. I’m greedy

They came to one of these conclusions, based on how they see themselves in the same position.

I can be of course one of these three in some other situations, but in the left pad incident, I done all purely with my heart, to stick to my values and principles which was behind my motivation to do open source for such a long time.

Ghazali is the best source I’m aware of about how to put heart in the driver seat of life, and without fighting but using logic, greed, anger etc as tools.

If my reference made curious about him, I couldn’t be more happy. Here’s a great lecture about his book, The Alchemy of Happiness: https://youtube.com/watch?v=zBwWc0DflRQ

installation fails for me

λ brew tap brandonyoungdev/tldx λ brew install tldx

==> Fetching brandonyoungdev/tldx/tldx ==> Downloading https://github.com/brandonyoungdev/tldx/releases/download/ curl: (56) The requested URL returned error: 404

Error: tldx: Failed to download resource "tldx" Download failed: https://github.com/brandonyoungdev/tldx/releases/download/v1...

I just got some good domains like mitte.ai and told.so a few months ago

Good summary of the paper: https://x.com/build__ship/status/1926930191185580176

Sorry, I don’t want to serve this group of users. Please download the image models and run them in your own hardware. Perfect for privacy and all that stuff

You can sign in with Google.

I had to close the sign up because there was so many abuse coming from regular sign ups.

'Sign in Google' is great because it eliminates low quality traffic who never pays and tends to be there for abusing the system.

https://mitte.ai — an AI image generator with focus on quality and details

so you can get logos / icons that doesn’t look AI generated.

it comes with Photoshop-like editor (https://mitte.ai/editor) so you can zoom into details and change / remove anything, or upscale, etc.

I built it for myself but now there’s good amount of paying users as well.

Incredible. So they're openly supporting genocide of Palestinians?

This is probably coming from the safety instructions of the model. It tends to treat the user like a child and don't miss any chance to moralize. And the company seems to believe that it's a feature, not a bug.

Yeah but the quality on Runpod is not reliable enough for productionizing it. Do you know a product that works reliably and it's built with Runpod serverless?

thanks, I'll reach out!

Did you consider other use cases in which people need custom models and inference other than just open source LLMs ?

I spent a month setting up serverless endpoint for a custom model last year with Runpod. It was expensive and unreliable, in addition to long cold boot times. The product was unusable even as a prototype, to cover the costs, I'd have to raise money first.

In a different product, I was given some Google Cloud credits, which unlocked me to put the product in front of customer. This one also needed GPU but not as expensive as the previous. It works reliably and it's fast.

Personally, I had two use cases for GPU providers in past 3 months.

I think there's definitely demand for reliability and better pricing. Not sure Fly will be able to touch that market though as it's not known for both (stability & developer friendly pricing).

P.S If anyone is working on a serverless provider and want me to test their product, reach me out :)

if you add fzf to speed up file / folder selection, you'll have your own llmcat :)

or https://github.com/azer/llmcat