Announcing: Voice Agents course and online community ...

@swyx and I are hosting a month-long technical deep dive into Voice AI and Voice Agents, starting in May. Our goals are to: ➡️ cover all the lessons we've learned over the last two years building realtime, conversational AI, ➡️ host fun sessions with all our favorite people who are doing related things, and ➡️ build a long-term online community. Sign-up link: https://lnkd.in/gnPuHyD4 We'll start announcing free credits for students on Monday. Sign up this weekend with promo code PHUNT for a super-secret Product Hunt community discount. Last year I signed up for the LLM fine-tuning course taught by Hamel Husain and Dan Becker. The experience was fantastic in every way. The material was great. The course expanded to cover way more than fine-tuning. It seemed like all of Twitter signed up. I met people in the course Discord that have become online and offline friends. Someone eventually dubbed the course "AI Woodstock." (I think credit for that goes to Swyx.) We think this is the moment to try to create a similar thing for voice AI. Voice interfaces are going to be a huge part of the near-future of computing. Voice agents are being deployed at scale today for a wide range of use cases. ➡️ collecting patient data prior to healthcare appointments ➡️ following up on inbound sales leads, ➡️ handling an increasing variety of call center tasks, ➡️ coordinating scheduling and logistics between companies, and ➡️ answering the phone for nearly every kind of small business. I'm personally excited about voice interactions for games, realtime video, and voice-enabled programming environments. https://lnkd.in/gnPuHyD4 Promo code: PHUNT
@OpenAI just dropped a 34-page guide on how to build AI agents. It’s full of great ideas but could be hard to navigate if you're non-technical. I made a simple, no-code breakdown of only what you need to know here and how to apply these ideas into the no-code AI agents & multi-agent workflows you build on @MindPal here: https://mindpal.space/blog/openai-ai-agent-guide-decoded-no-code

Here's my hacked-together, messy, voice-based dev environment:
Voice-driven loop with screen-shotting so the LLM in the loop can see what's in my terminal and editor. The prompt varies depending on what I'm trying to drive with this loop.
A few tool definitions that give read access to files and URLs.
A tool the LLM can send a block of output to that generates keyboard events, so the LLM can drive any editor/terminal.
A separate process watching a directory and constantly making LLM-driven git commits. (git autosave).
I have some pieces of this running most of the time. But I'm lazy, and doing other stuff, and I also try to use a variety of editors and tools, to see what's good lately. Which ... no stability, so my hacked-together stuff is always broken.
I don't want to replace @Windsurf / @Cursor / Claude code. A seriously good agent and expert-system dev toolkit is a lot of work.
I feel like Im using Youtube more than anything else every single day. I want to learn some "hidden gems" channels, cause I feel I seen it all. Ideally channels around solopreneurship/indiehackers, 0$ marketing strategies, how-to-sell stuff fast, philosophy and economics. Here are my GO-TO channels that I watch almost daily:
@Raoul Pal the Journey man (Macro investing in crypto)
@Money ZG (macro investing in crypto)
@Pursuit of Wonder (interesting concepts around societal psychology & philosophy)
@Fireship - Tech edgelord (lol)
@Greg Isenberg (Solopreneurship)
@Starter Story (Solopreneurship, How to build and market apps)
What are yours?
Until now marketers have had to guess what's a normal CPC, CPM or CTR in their niche. There's no real public information available. So we decided to fix this. Here's our public free Facebook Ads benchmarking tool.
It's based on real data, not some marketer's best guess from 2017. It's built on an anonymized data set of $2B+ of Facebook Ads ad spend. All data is fully aggregated and anonymized You can currently select CPC, CPM or CTR. And slice the data by industry, campaign type and country.
Is your CPM spike an industry-wide trend?
Is your CTR truly exceptional or just average for your vertical?
Hey everyone 👋
I’ve been building products for a long time (15+ years), and I recently tried using v0.dev for the first time. Honestly didn’t expect much, but I was surprised how quickly I got something real off the ground - not just a playground UI, but a fully working fitness app with protected routes, dashboards, flow logic, the works.
It’s called The HIIT PIT — and it’s live, but that’s not why I’m posting.
I’m more curious to hear from other devs and indie makers:
I'm super curious how everyone starts to vibe code? In the beginning I would simply jump into @bolt.new or @Cursor and just do a prompt and continue refining with the AI. I quickly realized this created a lot of issues as I didn't think about the structure, tech stack, and how I wanted the features to interact with each other and how the way I was building things would impact the user experience. I now do the following:
Write down a simple problem statement: "what am I trying to solve?"
Write down a simple solution statement: "what does the thing I'm building do (to solve the problem)"
Share the above with @ChatGPT by OpenAI and word vomit my thoughts, ideas, how I want the user to interact with my app, etc and ASK ChatGPT to turn everything I said and want into an easy to understand directive and instructions for an Engineer.
I then take the Engineer instructions and give it to a new chat in ChatGPT and ask it to turn those instructions into a prompt for an AI engineer and to break up the project into sections so that each time we focus on a section the app is shippable and keeps things easy to work on.
I take the output and paste it into my notes. I then give it to Cursor.
Once in Curosr, I create a new project folder and got at it!
Curious what everyone else does and if you've experience any things to avoid or must do