hey guys hopefully everyone had a pleasing day/evening.

If you're currently working on that "one last feature" I suggest you finish it up and get rested.

You have a great day full of challenges and mysteries ahead of you tomorrow and the best way approach it by having a good night sleep.

So if you're on social media scrolling, 120% stop that and instead listen to a podcast or watch a movie instead (best option: read a book).

DO NOT SCROLL BEZI SAET 🚫

I'm wishing you a pleasant rest, see you tomorrow 👍❤️

Thanks more moh❤

anytime ende

we truly and sincerely thank you from the bottom of our hearts for your efforts in giving us the opportunity to improve. Truly, keep up the great work, stay strong and keep moving forward.❤

Hohe ASR now ships with the server we run ourselves, and it is open.

Three commands and you have Amharic speech to text running on your own machine. No GPU, no account, no API key, nothing to sign.

git clone https://github.com/snapwre/hohe-serve && cd hohe-serve
docker build -t hohe-asr . && docker run --rm -p 8080:8080 hohe-asr
curl -F audio=@clip.ogg http://localhost:8080/transcribe

The image fetches the weights from Hugging Face itself and serves on ordinary processor cores. On a clean machine we timed it end to end: built from nothing, model loaded in 6.8 seconds, a 6 second Amharic clip transcribed in 1.1.

It streams too. Long audio is cut at its quietest moments and the text comes back in pieces as it works, rather than all at the end.

Also in the repository: exactly how we deploy it on AWS spot behind a load balancer, and the full speed measurements behind every number we have published, so you can check them instead of trusting them.

Model, CC BY 4.0: huggingface.co/snapwre/hohe-asr-amharic
Server: github.com/snapwre/hohe-serve
Try it without installing: t.me/dataset_ai_bot

Just letting everyone know we got gbi safe! Thanks for everything today, love you all!

This has never been more true than now thanks to AI

@selfmadecoder

Btw guys, I forgot to mention

There’s a mobile dev Intern position at our company right now. If you’re good at what you do, there’s a chance you can turn it into a full-time position.

If you’re interested, don’t miss it DM me with your cv @alnova19

@devwitheyob
#opportunities #intern

One of the biggest challenges I face:

when trying to watch educational videos on YouTube is spending hours watching them only to realize I didn't get what I was looking for.

Additionally, when a video has 2,000 or 3,000+ comments, it’s very difficult to read through them to understand what viewers are saying about whether the video is actually good or bad.

To solve this, I built VideoTrust AI Chrome extension that analyzes the video and tells you whether to watch it or not.

It reads through the comments, checks every detail, determines what the video is truly about, and assigns a Trust Score:

70 – 100: Watch it
45 – 69: Maybe
0 – 44: Skip

It explains everything clearly for each video. I've built and tested this extension. give me your feedback!

@Techyada21

💻 100 Days Coding Roadmap 🚀👨‍💻

📍 Days 1–10: Programming Basics 
– Choose a language: Python / JavaScript / C++ 
– Learn syntax, variables, loops, conditionals 
– Write basic programs & challenges 

📍 Days 11–20: Data Structures 
– Arrays, Lists, Stacks, Queues 
– Practice using built-in methods 
– Start solving problems on LeetCode or Codeforces 

📍 Days 21–30: Algorithms Fundamentals 
– Sorting: Bubble, Merge, Quick 
– Searching: Binary, Linear 
– Time & space complexity (Big O notation) 

📍 Days 31–40: Object-Oriented Programming 
– Classes, Objects, Inheritance, Polymorphism 
– Apply OOP to build small real-world projects 

📍 Days 41–50: Intermediate DSA 
– HashMaps, Sets, Linked Lists 
– Recursion, Backtracking basics 
– Solve 50+ problems for logic building 

📍 Days 51–60: Advanced DSA 
– Trees, Graphs, Heaps, Tries 
– Dynamic Programming intro 
– Participate in contests (CodeChef, HackerRank) 

📍 Days 61–70: Web Basics (HTML/CSS/JS) 
– Build portfolio website 
– Learn responsive design 
– DOM manipulation with JavaScript 

📍 Days 71–80: Backend + APIs 
– Learn Node.js / Django / Flask 
– Create REST APIs, connect with frontend 
– Use databases like MongoDB / MySQL 

📍 Days 81–90: Projects & GitHub 
– Build 2–3 full-stack apps 
– Use Git, GitHub, README files 
– Deploy apps (Netlify, Vercel, Render) 

📍 Days 91–100: Interview & Capstone 
– Revise top 100 DSA patterns 
– Mock interviews, resume prep 
– Complete one big project and publish it 

@Techyada21
#From_fb

"... በመስቀሉ ገነትን ከፈተ ፤ በመስቀሉ መድኃኒትን አደረገ ፤ መስቀል ከሁሉ በላይ ነው ፤ እርሱን ተሸክመን እንመላለስ ..."
እንኳን አደረሳችሁ

@Techyada21

Slop UI is the UI that AI cooks up that looks off, where it’s not supposed to be, doesn’t serve a purpose or when a UI is too generic

@Techyada21

Been thinking about this lots and I think these are the primitives you need to build any of the latest AI powered products (Chatbots, agents...)

At it's super core, you just need the interface, model and harness. Personalizations, autonomy and safety are more of an improvement/advancement layer than a core foundation, they mostly allow you to build a more agentic system.

So pick any interface you're comfortable working with (site, app, bot, CLI), pick a model router (ScholarXIV router, Vercel AI Gateway, OpenRouter), and then start building the harness — which is basically making the AI model work smoothly in the interface of your choice. Think of the harness as a workstation for the model selected, it gives it resources and tools.

The advancement layers are mostly compliance and better UX changes. The more secure the product is, the more time you'll spend using it and the more context aware it is the smoother the experience is. So safety and personalization are the cherry on top.

#Thoughts
@Dagmawi_Babi

Hey fam, Here’s the GitHub Tricks cheatsheet
Hope you find it useful!
share this someone who wants

@TechYada21
#github #gitTricks