DevEvent was sooo fun. We hiked, danced, played truth or dare, ate, and did so many random fun things. You guys have no idea, it was actually really cool😁

I also met some really great people, including Semer Nur. I had a really good time with him and learned a lot from his experiences. He was super humble and genuinely happy to help people, which I really respected.

Overall, I had such a great time. Good people, good memories, good vibes.

@devwitheyob
#TechVibe #DevEvents

Well, well, speaking of Cursor, SpaceX just bought Cursor.

@devwitheyob
#TechVibe #Cursor #AiAgents

Yoooo, 900 subs! You guys are the best. Thank you. You're gonna have a giveaway when we reach 1k. Also, let's make the community grow.

Join the discussion group: [Link]

@devwitheyob
#TechVibe #900subs

AAU about to make me crash out 🥲

One feature I think OpenAI should add is a popover-style “Ask about this” window inside long ChatGPT responses.

When I’m using ChatGPT to plan a project, it often gives me a detailed plan with 10 or more sections. While reading, I might want to ask a quick question about one specific part of the plan. Right now, I have to scroll back to the chat box, type something like “explain 4,” wait for the answer, and then scroll back to the original plan to continue reading. That gets annoying when I want to ask follow-up questions about multiple parts of the same plan.

A better experience would be, when I highlight or select a section of the response, it could show a small popover where I can ask a question about just that selected part. The answer could appear in the popover without disrupting the main conversation flow.

This would make it much easier to understand complex plans step by step without losing context or constantly scrolling back and forth.

@devwitheyob19
#OpenAI #features

Me watching Nahom Biruk and Solo Codes at this time while having an exam tomorrow😭. Great podcast, btw.

@devwitheyob
#TechVibe #podcast

One thing I relate to in the podcast was when Brook (Solo Codes) said he gives his clients access to tools like Lovable and other AI vibe-coding platforms, then tells them to simply type out what they want to build. He uses that input to understand their vision and then builds the right version for them.

Damn, that makes a lot of sense because it helps clients feel involved in the creation process. Instead of just handing requirements to a developer, they get to actively explore and express what they want for their company.

@devwitheyob
#TechVibe #podcast

Me installing tiktok to watch

nikatehilina vs johnny memes😂😂😂😂

I though I was the only one😁

Bullsh*t match I deleted tiktok right away😭

@devwitheyob
#TechVibe

It's time to get more aggressive about the life you want.

@devwitheyob
#TechVibe #random

Yooo, this is crazy. Two paychecks came in at the same time, one local and one on Upwork. Funny thing, I didn't expect them, I still have four or five days left. Life's feeling blessed 😁.

Sending one whole check straight to my mom. And guys, Grey works perfectly. now I'm waiting for the third job...

@devwitheyob
#TechVibe #random #upwork

And @CoolioCodes was actually super awesome ... he put everything together and organized it really well. Honestly, he might be the most extroverted person I’ve ever seen 😁

You guys should definitely follow him. He’s planning to make dev events a regular thing and do a lot more of them. Let’s get him to 1K by tomorrow too 🔥

@selfmadecoder

You guys, Claude Code is working in Ethiopia. I'm running my code without any VPN.

@devwitheyob
#TechVibe #ClaudeCode #AI

Compounding is the key

@devwitheyob
#TechVibe #podcast #tips

Finally shipped a new AI feature. At my current company, we have an LMS built for TVET centers around AA. My task was to build and deploy a RAG-based assistant inside the lms.

The assistant helps students understand topics from their LMS materials, but with guardrails so it doesn’t give direct exam answers or do assignments for them.

Tech-wise, I built a separate NestJS AI layer running internally on the same Docker network as the backend, with workers for indexing and re-indexing materials when they change.

Used Gemini embeddings, PostgreSQL, pgvector(for the vector store), and Drizzle.

The hardest part was deploying it without breaking the live LMS, keeping material indexing reliable, and making the AI layer secure. happy to see it live 🙌

@devwitheyob
#TechVibe #AI #feature

really nigga😁

@devwitheyob
#TechVibe #random

What is the one thing you fear the most at this time this is for the people who are in the hustle to get somewhere and become someone?

drop your ideas in the comments fam

@devwitheyob
#TechVibe #random