Ashaba Jasper

#firstpost 🎓 Excited to announce that I graduated with First Class Honors (4.62 GPA) in Computer Science from Uganda Christian University! A 🧵

The rich and the poor are both self made. But only the rich admit it.

FEAR EP out 11/14. NF.lnk.to/FEAR

Look at that everyone A LOCAL BUS !

whenever you feel behind in life just remember he didn't create the linux kernel until he was 21 years old

Adulting so bad that one unproductive day feels like failure.

Forgive us our trespasses as we forgive those who trespass against us

@code On point 3, you wouldn’t need to say ‘experience’ after UX , it already means user experience 😅.

🚀 v1.104 of @code is here! Check out what's new: 🤖 Improved coding agent integration 📄 AGENTS.md file support for better context 🔍 New Auto mode (Preview) for smart model selection 🔑 Model flexibility via BYOK extension API …and more: aka.ms/VSCodeRelease Here are the highlights 🧵

same reason, different results

my kind of rich

if you're not happy single, you won't be happy in a relationship. true happiness comes from closing 100 chrome tabs after solving an obscure programming bug, not from someone else

traits you see in successful people, but not normies.

Alcohol is the most harmful drug we use. It kills more people each year than wars, terrorism, car accidents, and homicides combined. Why do we cling to something so harmful? Watch our latest video to find out: kgs.link/alcohol

Make something people want.

Update to this slide 👍, a lot has changed since 2018–21. 1) “Common sense”: modern models score high on community tests (WinoGrande, CommonsenseQA). Proof: leaderboard.allenai.org/winogrande/sub… leaderboard.allenai.org/csqa2/submissi… 2) “Understand context”: today’s systems can read book-length inputs (1–2M tokens). Proof: developers.googleblog.com/en/new-feature… blog.google/technology/ai/… 3) “Learn from few examples”: few/zero-shot has been standard since GPT-3. Paper: arxiv.org/abs/2005.14165 4) “Needs huge task data”: pretraining is heavy, but in practice we adapt with tiny domain data via RAG + LoRA. Papers: arxiv.org/abs/2005.11401 arxiv.org/abs/2106.09685 5) “Explain decisions”: still hard, active research in explainable AI. Survey: arxiv.org/abs/2409.00265 Takeaway: progress is real (context, few-shot, efficient adaptation), but transparency/explanations remain open problems.

🤖 What AI Still Can’t Do (Yet) Despite rapid advancements, AI still struggles with: Common sense reasoning Humans still have the edge in many areas 💡✨ #AI #ArtificialIntelligence #FutureOfAI #Tech #Innovation

@fasc1nate You missed -oscopy where the surgeon grabs a camera and goes exploring.

The spectrum of empathy

seeing others achieve things should spark motivation not jealousy