Frustration-Log Micro-Tool Ideation
Beat the idea crisis: log a week of frustrations, then build tiny AI tools to kill them.
- Difficulty
- Starter
- Time to result
- ~days to results
- Steps
- 4
- Confidence
- 85%
A repeatable idea-generation method for the 'I have vibe-coding tools but don't know what to build' problem. For one week you deliberately track what you do and what frustrates you, look for recurring frustrations (yours and your teammates'), and build small AI tools to remove them. Use it in hackathons, for personal productivity, or to onboard a team onto AI.
Origin
Huyen describes an 'idea crisis': people have powerful build-anything tools but are stuck on what to make. Working with companies on hackathons, she and the team developed a guideline for generating ideas, centered on noticing weekly frustrations.
Core principles
- 01Specialization has left people without a big-picture view, which is why ideation feels hard.
- 02Frustration is a reliable signal of a buildable problem.
- 03Shared frustrations across a team point to higher-value tools.
How to run it
- 1
Log a week of friction
For a week, pay attention to what you do and specifically what frustrates you.
Pro tip Write frustrations down as they happen rather than trying to recall them later.
- 2
Interrogate each frustration
For each one ask: is there anything we can do about this? Could it be done a different way so it isn't frustrating?
- 3
Pool frustrations across the team
Have people swap or share their frustration notes; recurring, common frustrations are the strongest candidates.
Pro tip Common frustrations across multiple people signal broader value.
- 4
Build a micro-tool
Use AI/vibe-coding tools to build something small that addresses the frustration.
Pro tip Keep it a niche, single-purpose tool that just makes your life a bit easier.
In the wild
Lenny got frustrated that you can paste images into a Google Doc but can't easily get them back out ('Hotel California'). He vibe-coded an app where you give a Doc URL and it downloads all the images automatically.
→ Worked well, solved a recurring personal frustration, shared in the show notes.
Common mistakes
Starting from a blank canvas
Opening a vibe-coding tool and asking 'what do I want?' leads to paralysis; frustrations give you a concrete starting point.
Over-scoping
Trying to build a big product instead of a tiny tool that removes one specific friction.
Is it for you?
Best for
Individuals and hackathon teams who have AI build tools but no clear idea of what to make.
Not ideal for
Teams that already have a validated, high-conviction product bet and just need to execute.
From the transcript
“one tip is like go look from the last week right like for a week just like pay attention to what you do and what…”
“I do I'm very bullish on like using AI just create like micro tools like just something just like make your life a bit easier”
From the episode
Al Engineering 101 with Chip Huyen (Nvidia, Stanford, Netflix)