Been building more voice AI agents lately and their visual UX has been a fun design challenge to think through. While conversational capabilities have advanced (just check out the insane voice demos from Sesame, Deepgram, ElevenLabs, and OpenAI), choosing what users see

I feel this right now. I spend most my time trying to build the perfect context and tests that aren’t overly prescriptive and leave room for the agent to be better than me. I’ll often completely restart something I’ve been working on for hours because I think it’d be better to

Appreciate how Jean-Denis Greze 💡 frames context engineering as a search problem: "A correct answer is of little value if the agent has to 'brute force' the search with fifty expensive steps to find what a better tool could locate in two. [...] Designed well, great tools don't magically

“You can already see this shift in how people talk about “AI native” apps. The most interesting ones rarely start by building yet another end user UI. They start by sitting next to the warehouse, or next to the CRM, or next to the ERP, and then building opinionated agents that