Today, I’m officially sunsetting Avora. If you’ve found your way here from the old Avora domain, welcome, and thank you for being part of the project.

Avora started as a passion project during the early days of the LLM boom, when GPT-3.5 was just making waves. At the time, these systems were nothing more than chatbots. They couldn't actually do anything, nor had access to the web. So my goal was to bring a smart, context-aware chatbot directly into the browser. It honestly was just a fun experiment at first, but quickly became one of the most valuable learning experiences I've ever taken on. Building Avora not only taught me everything to know about integrating LLMs into apps, but also ignited my general interest in the field of artifical intelligence. I genuinely wouldn’t be where I am today without it.

Under the hood, Avora’s backend performed semantic search across your browser’s web content, then sent the most relevant, chunked context to the LLM. It was essentially RAG before RAG became a well recognized practice in AI integration. Seeing those ideas take off years later is kind of crazy.

While Avora’s chapter is closing, my work in AI and software is far from over. If you’re curious about what I’m building next, or just want to follow along, feel free to explore the rest of my site.

Thank ya'll for being around, and for being part of Avora’s story.