Welcome back to Tech Break by Friday, where we make big tech topics feel small and doable. In today’s episode, I’m sharing the story of how I joined a hackathon, used a tool called Lovable, and built an AI-powered research assistant — all in under 48 hours.
Even though I have a background in AI, I wanted to test how beginner-friendly vibe-coding really is. Using just natural language prompts (and no manual coding), I created a working MVP that connects with Zotero, fetches academic papers, and allows users to chat with their research library.
This episode is all about showing you that you can build your own AI project, even if you’re just getting started. I’ll walk you through:
The exact tools I used (Lovable, Supabase, Clerk, Zotero)
The problem I was trying to solve as a Master’s student
What the MVP includes (chat interface, Zotero sync, persistent history)
How you can recreate the project using prompts — no heavy coding required
And no, this episode isn’t sponsored. I’m just genuinely impressed with what Lovable made possible — and I think you will be too.
If you enjoyed this episode, please follow the podcast, leave a review, or share it with someone who’s been meaning to build something with AI but doesn’t know where to start. If you create something, please share it with me, either by dm or in the comments.
Thanks again for listening. I’ll see you next Friday, where we keep discovering new topics and ideas.
Blog of the full tutorial: https://open.substack.com/pub/kivroglouparaskevi/p/building-your-first-ai-project-beginner?r=56ehxy&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true
A special offer for podcast listeners: If you have a business and would love to implement AI and automation ethically in your workflow, you get two one-hour free consultations to make a strategy together. Book a call!
Sources:
Media:
Spotify | Instagram | Apple Podcasts | X
Share this post