You're Absolutely Right! is a weekly video show about AI tools, techniques, and the weird future we're all building whether we meant to or not.
Part Car Talk, part This Old House, part Pee-wee's Playhouse — every week, Scott and Justin open up the workshop and see what happens. Sometimes it's a product teardown. Sometimes it's solving a listener's problem live. Sometimes it's a conversation with someone doing something genuinely wild.
The name is a phrase people use when they're halfway dismissing someone — sycophant mode, polite disagreement, the thing you say when you're not actually engaging. We're trying to do the opposite: actually engage, actually try things, actually say what we think. Even when — especially when — we don't know what we're doing.
Nobody knows how to build with AI yet. That's the whole point. We're all professional beginners.
Scott and Justin each pick an AI product and take it apart. Not reviews. Not hype. Just: what does this thing make us think? What's interesting about how it works? What would we steal from it?
Listeners write in with their AI problems. We talk through it live, build a solution, and probably argue about the best approach. Like Car Talk but the car is your codebase.
The weird tricks, workflows, and superstitions we've picked up working with AI every day. Yours too. Send us the thing you do that you can't explain but can't stop doing.
Demos of what we've been building. New products, experiments, things that work on our machines and might work on yours. No promises.
We bring someone on who's doing something absolutely unhinged with AI. WebSim, WorldSim, things that make you question what software even is anymore.
Quick hits on what caught our attention this week. New releases, shifts in the landscape, things that made us say "huh" out loud.
Scott runs Sublayer, an AI-native software studio in New York. He writes the weekly newsletter Works on My Machine (about what happens when nobody knows how to build with AI yet) and Near Zero (speculative fiction about a world where the cost of software approaches zero). He thinks Ruby might be the future of AI agents and has been wrong about enough things to be interesting.
Justin is a product thinker, tool explorer, and the other half of the workbench. He brings the "did you actually try using it though?" energy that keeps the show grounded. When Scott is off theorizing about cybernetics, Justin is the one who opens the product and sees if the button works.
New episodes every week on YouTube.
Subscribe on YouTube