Good Inside: What If a Parenting Book Could Talk Back?

Becoming a father broke my confidence. I had no idea what I was doing, and every parenting forum on the internet made it worse. Then a friend handed me “Good Inside” by Dr. Becky Kennedy, and something clicked.

The core idea is disarmingly simple: your kid is good inside, even when they’re screaming at you. Tantrums, hitting, defiance… these aren’t signs of a bad kid. They’re a child struggling to express something they don’t have words for yet. Dr. Becky’s reframe hit me hard: your job isn’t to fix behavior. It’s to build connection first. Reduce shame. Tell the truth. At 2 AM, mid-meltdown, with your patience gone, none of this comes naturally. But knowing the framework changes how you respond.

The book covers everything. Hitting, sibling rivalry, separation anxiety, picky eating, lying. All grounded in one principle: connection first, behavior second.

Here’s the thing that kept nagging me: this book has 29 chapters. No sleep-deprived parent is going to flip through an index while their toddler is mid-tantrum. The wisdom is there, but the access pattern is broken.

So I started thinking: what if you could just describe your situation, “my toddler won’t stop hitting his sister and I’m losing it,” and get back a Dr. Becky-style response? Not some generic parenting tip from a search engine. Something grounded in the actual frameworks from this specific book, tailored to your specific moment.

LLMs are just barely capable enough to make this work. ChatGPT had been out for about a year when I started thinking about this. People were experimenting with RAG pipelines and domain-specific knowledge, but almost nobody was applying it to parenting. The idea of an AI coach informed by one specific, evidence-based philosophy felt both novel and genuinely useful.

Picture it: a parent types in plain language what’s happening. An LLM, grounded in Good Inside’s principles, responds like a patient coach. Not a replacement for reading the book. A complement. An interactive way to access its insights exactly when you need them, in the exact context you need them.

Most people still think of AI as either a chatbot or a code generator. Pointing it at a specific body of knowledge and getting reliable, thoughtful advice back? That’s still an unconventional idea. RAG pipelines are clunky. Prompt engineering is more art than science. But there’s something genuinely exciting at the intersection of AI and parenting.

Here’s my contrarian take: most “AI for parents” products are solving the wrong problem. They’re building generic advice engines, basically a fancier Google search. The real opportunity is AI grounded in a specific philosophy, one you already trust. The value isn’t in breadth. It’s in depth and consistency. A parent doesn’t need 50 conflicting opinions at 2 AM. They need one trusted voice, available on demand.

I’m not sure if I’ll ship this. But the idea is exciting. Building tools that meet people where they actually are, not where a textbook assumes they’ll be. That’s the kind of thing worth building.