Ivy - Bringing LLMs to 35 Million offline students in Ethiopia

Source: Dev.to

Overview

I’m writing this from Addis Ababa. While the world is talking about the latest LLM cloud features, 35 million students in Ethiopia are being left behind because they simply don’t have stable internet or the hardware to run “modern” education.

I’ve spent the last few months building Ivy, a specialized offline tutor designed to run on low‑end Android devices with zero connectivity. It’s an architecture built on edge‑inference and local language support (Amharic) so that a kid in a rural village has the same educational “co‑pilot” as someone in London or New York.

I’m a solo founder and I’m currently 95 % of the way through a global AWS challenge to get this project the resources it needs to scale. I’ve reached 50 likes on my own, but I’m at a point where I need the support of the broader tech community to reach the Top 50 and secure the next round of funding/support.

If you believe that state‑of‑the‑art education should be a tool for all students—not just those with a fiber connection—I would be incredibly grateful for your support.

Technical Highlights

• Edge inference: Runs entirely on the device, no internet required.
• Local language support: Fully functional in Amharic.
• Low‑end hardware: Optimized for inexpensive Android phones common in rural Ethiopia.

How You Can Help

• Click the link to my official AWS entry:
• If you find the “Why” and the “How” compelling, please hit the “Like” button on the entry.

I’m happy to answer any questions about the technical hurdles of edge‑inference or the reality of building tech in Ethiopia. Thank you for reading.