We tested an AI story chat on a platform with 4M+ MAU. People who used it spent 100 minutes on average. But only 10% tried it. So the feature worked and distribution didn't. A/B tests tell you which problem you actually have.
Bonaca is an AI tool that tells yacht charter companies when to lower a price. The first thing every user asked was “why this price?” So I designed one screen for that question: tap it, type it or say it, and you get a one-line answer instead of a table to read.
I wanna know the most UNHINGED way you're using AI. And I mean truly unhinged.
I'm trying to seriously level up my AI skills and I feel like I've barely scratched the surface of what's possible
I'm not talking about cute little ChatGPT prompts or asking AI to write your emails.
I wanna hear about the crazy workflows, agents, automations, and things you've built that make you go wait you can actually DO that?!
Especially the stuff most people don't even realize is possible.
What tools are you using? What have you automated? What are you making AI do that would've sounded insane two years ago?
And if you know of any good bootcamps, courses, creators you follow, or other resources to learn this stuff, PLEASE share.
Not unhinged, but I got my database and Claude talking to each other, and now they gossip about my stock levels behind my back. It's an MCP server: instead of digging through "sales_report_FINAL_v3_really_final.xlsx", you just ask "which items are low on stock this week?" or...
Building a Voice AI Assistant That Turns Phone Calls Into Thoughtful Follow‑Through
I built a Django backend for a family-focused AI phone assistant. A registered caller connects through Twilio, and Django Channels streams the call to an OpenAI Realtime session for a natural, two-way conversation.
The backend adds caller-specific context, selected voices, subscription minute limits, and interruption handling. After the call, background tasks generate a separate summary. A follow-up email is sent only when requested and the recipient is confirmed.
I also built APIs for managing relations, subscriptions, and dashboard usage data. One of the most interesting parts was connecting the whole lifecycle: incoming call, real-time conversation, usage tracking, and post-call follow-through.
Sending the follow-up email only after the recipient is confirmed is a careful rule: a wrong name heard on a call could send it to the wrong person. How does the confirmation work: does the assistant repeat the name during the call, or does the caller confirm later in the dashboard?