JJ

We didn’t build Fluenty first. We started building voice agents for businesses. The agent part worked much faster than expected. Then came the real problem. Every client asked the same question: “ok… where does this live on my website?” Telegram demos worked. Playground demos worked. But a real business site needed something else: • branding • mobile behavior • business hours • a way to know if...

One thing we love about @Retell AI - Voice AI Agent is that it solves the hard part really well: building voice and chat agents that actually work. Getting an agent live isn’t hard anymore. What we kept seeing, though, is that once the agent is on a real website, teams start asking a different set of questions. Not “does it work?” but: How do users enter the conversation? How do chat and voice...

When we first launched prjct/cli, the idea was to work with every AI editor — @Claude Code, @Cursor, @Windsurf (my fav), @Codex by OpenAI. Sounded great, right? But in practice it was a mess: Tons of extra code just for compatibility Features blocked because we had to play lowest-common-denominator Testing was impossible to trust So in v0.5.0 we pivoted to @Claude by Anthropic. That decision...

We launched prjct-cli v0.1.0 not long ago, and today we’re already at v0.3.0 I want to thank everyone who took the time to send feedback, open issues, or share thoughts. It made a real difference. Here’s what’s new: Copy & Docs Corrected the command count (18, not 19) Removed non-existent commands like /p:git push or /p:roadmap add Added 5 real commands that were missing from the docs Unified...

Every time I tried to turn an idea into a roadmap, I’d jump between Claude and ChatGPT. One gave me X, the other Y, and nothing lined up. Cursor was better, but not exact (and pricey). Claude Code felt closer, but I like being in control. For me, AI should be a copilot — not the driver. Context windows filled up fast, and I’d lose the thread. With prjct/cli, I finally get that control back....