Notes

All notes

22 pieces

  1. 2 Nov 2026

    Note

    Diverse by prompt, biased by role: how AI illustration passes society's stereotypes on

    I asked for more diverse people in the sketch notes an AI pipeline draws for me. The AI writing the scene descriptions obliged by attaching an identity to each scene, and in doing so gave a wheelchair user a stage to conduct from, and the cleaning to an older Black man, alone, scrubbing a heap. The image model drew exactly what it was told, so every check passed.

    Designing for trust · Knowing if it's any good · Checking the checkers

  2. 30 Oct 2026

    Build

    EXP-009 Mullo

    The best page, the worse journey: when an AI tool's quality isn't worth the effort

    Checking the checkers · Designing for trust · Knowing if it's any good · Seeing the cost

  3. 30 Oct 2026

    Guide

    How I evaluate AI output

    A living guide. I wrote v1 in 2025 after building Launchling's evaluation framework; v2 is how I'd do it in October 2026. I've kept both, because the gap between them says as much as either version does.

    Knowing if it's any good

  4. 29 Oct 2026

    Build

    EXP-007 This lab, rebuilt

    Nothing goes live unread: building a public site with AI agents, without losing control

    Building with agents · Checking the checkers · What it knows and can reach

  5. 29 Oct 2026

    Note

    A recipe for honest pictures: how to avoid AI image slop without copyright trouble

    People choose what to cook by picturing it, but cookbook photos aren't mine to reuse, and AI pictures of dishes nobody has cooked are usually wrong. Here's how separating looking from drawing gave every recipe in Fold a picture that's true to the dish without copying anyone's.

    Seeing the cost · Checking the checkers · Designing for trust · Building with agents

  6. 24 Oct 2026

    Build

    EXP-006 Fold's review panel

    Never mark your own homework: independent AI reviewers for a solo builder

    Checking the checkers · Breaking it on purpose · Designing for trust · What it knows and can reach

  7. 15 Oct 2026

    Build

    EXP-005 Fold

    Cheap to build, hard to stop: when AI builds faster than you can learn

    Is it worth building · Seeing the cost · Designing for trust

  8. 13 Oct 2026

    Build

    EXP-004 emilysas.com

    The bottleneck moved: where the time goes when an AI agent writes the code

    Building with agents · Seeing the cost

  9. Jul 2025

    Migrated 2025 post

    Satirical Startups & French Pirate Founders: How Launchling Handles Absurdity by Design

    How Launchling handled absurd, manipulative and unethical startup ideas when I tried to break it on purpose.

    Breaking it on purpose · Designing for trust

  10. Jul 2025

    Migrated 2025 post

    How I Used AI Pair Programming to Build a Robust Evaluation System (Without Vibe Coding)

    The structure I gave ChatGPT and Cursor to build Launchling's evaluation pipeline, shortly before it stopped completing runs.

    Building with agents · Knowing if it's any good

  11. Jul 2025

    Migrated 2025 post

    From MVP to Momentum: A Week of Big Leaps for Launchling

    A week of Launchling changes: new pages, working email, the Risk Mapper and real test coverage.

    Is it worth building

  12. Jul 2025

    Migrated 2025 post

    Post-Launch Polish: What I’ve Been Improving on TinyLaunch

    The fixes I made once real people started using TinyLaunch, mostly for mobile, and how I split work between ChatGPT and Cursor.

    Is it worth building

  13. Jul 2025

    Migrated 2025 post

    From TinyLaunch to Launchling

    Why I renamed TinyLaunch as Launchling, and who I meant it for.

    Is it worth building

  14. Jul 2025

    Migrated 2025 post

    From Side Project to Stress Test: Using AI to Evaluate TinyLaunch

    My first attempt at scoring TinyLaunch's plans with GPT-4o, on 11 real plans, and what the scores told me to change.

    Knowing if it's any good

  15. Jul 2025

    Migrated 2025 post

    TinyLaunch Update: From Framer MVP to Full React Build — Lessons, Challenges & Next Steps

    Why I moved TinyLaunch from Framer to a React app, and the Vercel bug that took longest to find.

    Building with agents

  16. Jul 2025

    Migrated 2025 post

    TinyLaunch Is ‘Launched’: From Framer to Fully-Coded Product

    What TinyLaunch could do after the React rebuild, its two-step GPT-4o prompts, and how I split work between Cursor and ChatGPT.

    Is it worth building · Building with agents

  17. Jul 2025

    Migrated 2025 post

    Why I Built TinyLaunch – Helping Tiny Ideas Take Flight

    Why I started TinyLaunch (later Launchling): the problem, what I wanted to learn, and the first no-code alpha.

    Is it worth building

  18. Jul 2025

    Migrated 2025 post

    Trying (and Failing) to Send Emails from My Domain

    Trying to send a welcome email from my own domain through Zapier and Namecheap, and the fix that finally worked.

    Building with agents

  19. Jul 2025

    Migrated 2025 post

    Building My AI Product Portfolio Site with Notion + Super

    How I set up the first emilybuilds.ai on Notion and Super.so in about two hours, and what was fiddly about it.

    Is it worth building · Seeing the cost

  20. 6 Jul 2025

    Build

    EXP-001 emilybuilds on Notion + Super

    Deciding not to build: when no-code is the right call, and when it stops being one

    Is it worth building · Seeing the cost

  21. 1 Jun 2025

    Build

    EXP-002 Launchling

    All tests passing: when checking an AI product outgrows the product

    Knowing if it's any good · Seeing the cost · Is it worth building

  22. 1 Dec 2022

    Build

    EXP-000 Ambient care-plan scribe

    Before care had scribes: drafting care plans with AI in 2022

    Designing for trust · Is it worth building