codewithstefan's profile picture. Follow me on my path to 10k MRR (still progress: 0€ ....)

OrthodoxCoder

@codewithstefan

Follow me on my path to 10k MRR (still progress: 0€ ....)

مثبتة

I made 1k MRR in 2 Months #SaaS #dream (Screenshort as proof) (Haters will say its fake)

codewithstefan's tweet image. I made 1k MRR in 2 Months 

#SaaS #dream

(Screenshort as proof)

(Haters will say its fake)

Devlog #12 cleaned up error handling: dropped a redundant try/catch, added retries (1s, 2s, 4s) for openrouter requests, covering api + json errors. also typed the response properly 🚀


added error handling instead of a simple 5 line fetch request its now a function with over 50+ lines just for requests to open router


Devlog #11 refactored overwrite reduced good amount of lines in code TODO: better error handling catch em all & on fetch from openrouter retry


Devlog #10 removed logs added script for downloading images added other logs for debug 😂 added k6 load testing script burned some money testing AI Downloaded 1001 images 🔥


i downloaded 1001 images now

codewithstefan's tweet image. i downloaded 1001 images now

encountered some unexpected errors, while load testing, like how can it miss hash check??? I have to find a method to load test my api without burning through credits, I'll probably have to download like 100+ images and set up some tests i can execute for free


Devlog #9 Implemented async functionality - This included rewriting some logic to get the required credits earlier renamed a type implemented express error handeling Tomorrow refactor handeling for overwrite, new and missing translations because overwrite and new are basically…


1. While using google (if google says stop openAI can't do anything) 2. Siri is native and can be extended with a simple API call 3. based on chromium and google... ...

How OpenAI dominates the next 20 years: 1. Web app (beats Google search) ✅ 2. Mobile app (beats Siri) ✅ 3. Browser (beats Chrome) ✅ 4. Operating system (beats MacOS/iOS) ◻️ 5. Device (beats Apple) ◻️ If they keep executing...



Devlog #8 migrated everything to prepared statements added user type (JSDoc) continued async/webhook feature ⚙️ added more error handling todo: finish getMissingTranslations() + fix error


I one shot refactoring multiple sql queries without AI 😎


I also started playing around with Hoppscotch because Postman didn't work today...


Devlog #7 ✨ Implemented tags — you can now send them in the API request as a string array 🏷️ ⚙️ Added async + webhook validation 🔄 🧰 Improved error handling (no more ugly responses) 🛠️ npm found 1 high severity vuln — fixed! ✅ 🧠 Added Sentry for error monitoring 👀…


LLM's should get faster by time? Right????


😭😭😭😭😭😭

you can't make this stuff up

alxfazio's tweet image. you can't make this stuff up


excited ngl 🔥

This week will be super hard or super simple. Nothing in between. Huge releases and cloud infra improvements.



I also benchmarked grok-4 fast and gemini 2.5 flash on @OpenRouterAI My application in some cases is speed sensitive, so i debuged the responsetime with postman. The resoning tokens make grok 4 lose hard 6809 ms compared to 1925ms on average


Devlog #6 alt_text now returns the first requested language 🌐 asset_id saved to DB + deduped per user 🗂️ Fixed quote-escaping bug 🐛 Testing API with picsum.photos 🧪 works amazing btw TODO: add tags to API tomorrow 🏷️


picsum.photos/seed/<randomseed>/600/800/<,.png,.jpg,.webp> Is amazing and helps me test my app, because i need random images


United States الاتجاهات

Loading...

Something went wrong.


Something went wrong.