Passion-Project Learning (An Eval for Your Life)
Learn new tech by picking one passion project that forces you to touch everything you need to learn
- Difficulty
- Easy
- Time to result
- ~weeks to results
- Steps
- 3
- Confidence
- 85%
A method for staying technically sharp amid rapid AI change: instead of passively reading or listening, pick a personal project you genuinely care about that happens to touch a wide span of skills you need to learn. Passion pulls you deeper, and the project acts like 'an eval for your life' — a concrete test that verifies you actually learned. Especially recommended for people who don't know where to start with AI.
Origin
Oji Udezue's personal practice for staying hands-on; Ezinne uses the same advice when coaching AI newcomers. Framed against 'humility is teachability, and teachability is survivability.'
Core principles
- 01Humility is teachability and teachability is survivability — with no blueprint, you must keep learning hands-on
- 02Don't just read, listen to podcasts, or pontificate — actually build things
- 03Code is now 'essentially architecture and English,' so writing it is far more accessible than a decade ago
- 04Pick a project that touches many skills you need, so passion drives sustained learning across all of them
- 05Anchor the project to something you are genuinely passionate about — fashion, reading, your home — so motivation doesn't fade
How to run it
- 1
Adopt a learner's stance
Go back to 'day one' regardless of seniority. Accept being taught by more junior people and sign up for beginner AI courses even as a CPO or principal PM.
Pro tip Humility is teachability; teachability is survivability — treat learning as career survival, not ego risk.
Watch out Assuming you already know product because you have experience is the mistake that leaves senior people behind.
- 2
Pick one passion project that spans many skills
Choose a project you genuinely care about that forces you to touch a broad set of things you need to learn, so intrinsic motivation carries you through each new skill.
Pro tip Frame it as 'an eval for your life' — a concrete build that proves you actually learned, not just consumed content.
Watch out Generic 'I'll go play with Lovable' rarely sustains learning; a specific problem you're motivated to solve does.
- 3
Get hands-on across the full stack of the skill
Convert PRD writing into prototype writing, write and call APIs yourself (e.g. with Postman), subscribe to the AI tools to learn what each model is good for, and build MCP servers connecting real systems.
Pro tip Treat tool subscriptions as 'a cost of learning' so you understand each model's strengths first-hand.
In the wild
Oji has spent years automating his home. He built a 'super sensor' that sees, hears, and senses humidity and temperature, running Home Assistant with fine-tuned and large LLMs plus small Whisper voice models. The house precools kids' rooms before they return and adapts to occupants. Building it forced him to learn quantization, fine-tuning, MCP servers, and local model deployment.
→ A passion project became a broad, self-verifying curriculum in modern AI engineering.
Ezinne coached an older woman who didn't know where to start with AI but loved the 'outfit of the day' idea. She catalogued all her tops, bottoms, dresses, and shoes and built an LLM project that recommends an outfit based on the day's temperature.
→ Anchoring to a genuine passion gave a non-technical person a real, usable AI build and an on-ramp into the tools.
Common mistakes
Staying a pontificator instead of a builder
Reading newsletters and listening to podcasts without building leaves you 'a pontificator guy in the clouds' who hasn't actually learned the tools hands-on.
Picking a project you don't care about
Without genuine passion, motivation fades and the learning stalls; the project must be anchored to something you actually want, so the desire keeps pulling you through new skills.
Is it for you?
Best for
Experienced professionals and non-technical people who feel behind on AI and need a durable, motivating way to get hands-on
Not ideal for
People who need a specific production skill on a deadline and can't afford an open-ended exploratory project
From the transcript
“humility is teachability and teachability is survivability”
“I've written more code in the last one year than I have in the last 10 years because code is now essentially architecture and English”
“I'll take a pick a project that touches a lot of the things that I need to learn that I'm interested and passionate about and…”
“having a very specific problem to solve for yourself is a really good way to build and and get into these tools because you're motivated…”
“she started a project where she actually had she took the time to take all her tops, all her bottoms, all her dresses, all her…”
From the episode
How AI is reshaping the product role
Oji and Ezinne Udezue