This weekend was a good reminder that there’s still so much to learn in AI, even when you think you already know the basics. I started with the AI Residency Cohort 12 onboarding session by Decoding Data Science, where Mohammad Arshad walked us through the evolution of AI, how models have developed over time, and what we’ll be learning and working on during the next 20 weeks. We also explored concepts like tokenization, how language models process text, and AI hallucinations — why AI can sometimes give answers that sound completely confident but are actually wrong. It made me think about how important it is to understand what happens behind the scenes instead of just knowing how to use AI tools. Later, I attended the Agentic AI Engineering Masterclass, where we explored four layers of building agentic AI systems: • Prompt engineering • Context engineering • Harness engineering • Loop engineering One thing that really stuck with me was that getting a good response from AI is only one part of the process. Building a system that can use tools, check its results, and try again when something goes wrong is a different challenge altogether. And a little highlight… 🏆 I came 1st in the Kahoot quiz! 😂 Definitely a fun moment during the session! As someone studying IT with a focus on AI, these sessions make me even more curious about how AI systems are actually built and what we can do with them beyond basic prompting. Now I'm looking forward to the next 20 weeks of the residency — learning, experimenting, making mistakes, and hopefully building things I'm genuinely proud of. Thank you, Mohammad Arshad and Decoding Data Science, for the sessions and the opportunity to learn! One weekend, two learning sessions, and so much more to explore. Here's to getting better, one step at a time. #AI #AgenticAI #AIEngineering #DecodingDataScience #AIResidency

Really excited to be part of the Decoding Data Science (DDS) event at AI Everything Abu Dhabi! 🤖 Looking forward to meeting fellow AI enthusiasts, learning from the community, exploring new ideas, and getting hands-on with what’s happening in the AI ecosystem. It’s always exciting to step outside the classroom and be around people who are actively building, experimenting, and learning with AI. Can’t wait for this! 🔥 #DecodingDataScience #DDS #AIEverything #AbuDhabi #AI #ArtificialIntelligence #Learning #Community

Really happy to share that I’ve been selected as one of the Top 35 finalists in the Decoding Data Science community. What makes this exciting for me isn’t just the selection — it’s knowing that the projects, experiments, and learning I’ve been putting in are actually moving somewhere. DDS has pushed me to go beyond just learning concepts and actually build, test, break, and improve AI applications. Now onto the next stage. 🚀 Grateful for the opportunity, and even more motivated to see what I can build from here. Top 35 → next goal: keep going. 🔥 #DecodingDataScience #DDS #AI #LearningByBuilding #AIEngineering

Building an AI application is much bigger than simply choosing an AI model. The more I learn and build, the more I realize that the real engineering happens around the model. I’ve started asking questions like: → What data is the system receiving? → How should that data be processed? → What should AI handle, and what should traditional code handle? → How do we validate the output? → What happens when the model is wrong? This has changed the way I look at AI projects. Instead of thinking: “Which AI model should I use?” I’m learning to think: “What is the best system I can build around the problem?” That shift in thinking is probably one of the most valuable things I’ve taken away so far from Decoding Data Science. Still learning, building, testing, and occasionally breaking things along the way. 😅 #DecodingDataScience #DDS #AI #ArtificialIntelligence #AIEngineering #LearningInPublic #TechCommunity

For a while, my thinking was: “What cool AI thing can I build?” Now I'm trying to reverse it: What annoying problem can I find that AI can actually solve? That small change makes a HUGE difference. Because when I start with the technology, I end up with: 🤖 “AI chatbot for ___” 🤖 “AI assistant for ___” 🤖 “AI-powered platform for ___” …and honestly, a lot of these are just the same idea with a different UI. So now, before I build anything, I'm asking: What is the actual problem? Who experiences it? How are they solving it now? Why is the current solution bad? Does AI genuinely make it better? Would someone actually use/pay for this? I'm still figuring this out, but I think this mindset is going to save me from building 10 useless AI apps just because I can. 😂 How do you decide whether an AI idea is actually worth building?