# 当我们什么都能做，还知道自己想做什么吗？When We Can Do Anything, Do We Still Know What We Want? — Ryo Lu, ex-Cursor

Ryo Lu

[Talk page](https://ai.engineer/talks/y9Sg--RDtks-when-we-do-anything-do-we-still) · [Video](https://www.youtube.com/watch?v=y9Sg--RDtks)

## When We Can Build Any Journal, Which One Do We Want?

Former Cursor founding designer Ryo Lu uses an AI-built journal to show how editing a single feature can reveal what you want, where a tool should give you control, and what someone else sees in the same design.

### What happens after the first moment of amazement

[0:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=3s)

The host introduces Ryo Lu as a founding engineer at Cursor, then corrects himself. Lu was a founding designer. At AI Engineer Shanghai, Lu opens with a reaction familiar to people who make software. Someone turns a sentence into a website, builds a feature in minutes, or wakes up to a night’s worth of work that includes things they never thought to ask for. Lu knows the excitement of finally being able to make something he could previously only imagine. He also knows the thought that comes a beat later. Is he falling behind again?

A skill that took years to acquire can suddenly look like a sentence anyone could type. Learning another tool or starting a few more agents might help him keep pace, but there is no obvious point at which he would have done enough. Lu spent part of his career designing collaboration tools so people could build their own systems, then tools that helped people make software. He still wants to close the distance between imagination and reality. What concerns him now is whether, as that distance shrinks, the person doing the imagining gets any better at deciding what is worth making.

### An Apple computer Lu wanted to understand

[3:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=183s)

Lu traces some of the anxiety to a habit of using a capability as proof of identity. An engineer might take pride in writing code by hand. When a machine can do that work, the worry reaches beyond employment to a more personal question about what made the engineer distinctive. Lu goes back to an encounter with a computer before he had a professional identity to defend.

After moving to Canada as a child, he saw an Apple computer whose screen seemed to float. He couldn’t afford one, so he changed the icons and fonts on his own computer and worked out how to reproduce the glossy buttons that looked good enough to lick. The details mattered because of the feeling the computer had given him. Trying to produce that feeling himself gave him a reason to look more closely.

![A blue desktop shows a window of preference icons overlapping a news window with the Finder face. An Audio CD icon sits at upper right and the Dock runs along the bottom.](https://ai.engineer/api/talk-essay-media/196ce548a90983931c4eeea135747850c256abb4821edbed0259411b53f3fcd0)

Mac OS X desktop with overlapping windows

Earning a living at design brought him colleagues who cared about such things. It also turned play into output and personal taste into decisions made on other people’s behalf. He grew practiced at estimating value and answering when something could ship. He wants to preserve the earlier impulse, too. A machine might produce ten versions almost immediately, and he might still want to spend time fixing one detail in one version. Often he can feel what bothers him before he can say why. That interval is part of the design process he worries a faster tool might skip.

### Two minutes from a loose request to a daily routine

[5:46](https://www.youtube.com/watch?v=y9Sg--RDtks&t=346s)

Lu introduces the journal with an imagined exchange. Suppose he asks an AI, “Make me a tool to record my day.” Two minutes later, it tells him the journal is ready. Lu shows the working journal along with the AI’s reasons for its features. Mood scores will help him spot trends. Daily reminders will help him establish a habit. Weekly reports will give him an occasion to reflect. The journal also tracks writing streaks. Each addition sounds sensible enough that Lu can picture himself approving the whole thing.

Consider what the AI had to decide to get there. A mood score makes part of a day comparable to other days. A reminder establishes a time to write again. Tracking a streak makes a missed entry consequential, and a weekly report gives reflection a schedule. Lu had asked for a way to record a day. He had yet to decide whether he wanted to score his life or undertake to write every day.

An obvious error would give him something to challenge. A coherent answer is harder to examine because the rationale arrives before he has formed an opinion. He may well prefer what the AI made to anything he would have devised alone. His concern is what he would practice if every project went this way. He would get many chances to judge whether a rationale sounded reasonable and fewer chances to develop one from his own experience.

### Which feature would you take out?

[7:32](https://www.youtube.com/watch?v=y9Sg--RDtks&t=452s)

Lu asks for a show of hands on four possible cuts, one at a time. Who would drop mood scores? Writing streaks? Daily reminders? Weekly reports? He offers his own answer. He would probably remove the streaks, reminders, and reports. He can already imagine missing a day and feeling as if he owed somebody an entry. He wants to be able to record something that happened to him without acquiring homework.

![Three red X marks cross out the journal’s six-day streak, 21:00 reminder and weekly report. The selected date, four filled mood stars and three-line entry remain visible.](https://ai.engineer/api/talk-essay-media/e74cdd1f5d3a541b463815735f8fcfc56bb8e3982ef20e4cd64a4ea45affddf5)

Streak, reminder and weekly report crossed out

Another person might depend on the reminder or enjoy keeping a streak alive. A third might want to abandon the daily structure altogether and arrange fragments of life on an infinite canvas. Asking what features were missing would have kept the discussion inside the AI’s original conception of a journal. Choosing something to remove lets each person examine one consequence of using it. Lu didn’t need a detailed specification to get started. He needed something concrete enough to provoke a preference, and a moment to work out where that preference came from.

### Two ways to revise the same journal

[9:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=554s)

Lu is happy to hand off a job when he already knows its purpose. If he asks a tool to move data or convert a file, getting the finished result may be all the involvement he wants. Revising the journal is different because trying the change may alter his understanding of the job. He compares two screens handling the same request.

The screen on the left works like a service counter. It processes the request and hands back a new version. If Lu likes it, he can use it immediately. But suppose the revision removes a reminder he liked, or changes the design along with the features. His choices are to accept the result, reject it, or explain again which parts he wanted to keep. Even a small correction requires him to reconstruct his intent for the tool.

The screen on the right works like a workbench. Before execution, Lu can see the proposed changes, including the features marked for removal and the tool’s suggestions. Seeing the reminder slated to disappear might be the first thing that makes him realize he wants it. He can intervene in the plan then. After execution, a live preview gives him a different opportunity. He might try a layout of his own and discover, only on seeing the result, that it doesn’t work.

![On the left, a user asks to remove a streak count and objects when the reply announces extra features. On the right, a journal preview offers four feature choices, with A highlighted beside Mood score.](https://ai.engineer/api/talk-essay-media/f471e361c82ee28a98d74620d5282107e1e23c81014942dcdb5aba0035bcc006)

Removing the streak count and choosing journal features

For Lu, files, plans, edits, and differences between versions all offer places to step in. A person interested in one decision should be able to find it, understand why it was made, and change just that part. Reading every line of code would be an unreasonable price for keeping one reminder. Chat can belong on a workbench, too. Lu judges the interface by what the person can inspect or take back on this particular task.

### Three design principles for tools that people can steer

[12:29](https://www.youtube.com/watch?v=y9Sg--RDtks&t=749s)

First, expose important assumptions, make each change small enough to try, and let people reclaim work they delegated. In the journal, listing the reminder among the proposed deletions makes an assumption visible at precisely the point when someone can do something about it.

Second, offer choices within the answer. A confirmation button at every step would give the user plenty to do while still asking much the same yes-or-no question. The useful choice here is being able to keep the reminder while allowing the other edits to proceed.

Third, remove obstacles while preserving the opportunity to form a judgment. Having to explain the entire journal again is an obstacle. Trying a layout and noticing what feels wrong is valuable work. A live preview can shorten the path to that experience and still leave the discovery to the person using it.

### Choosing between assistance, a rough draft, and tea

[13:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=786s)

Even at a well-designed workbench, Lu still has to decide when to sit down. Email offers a familiar precedent. Some things are worth putting in writing, some deserve a face-to-face conversation, and some can wait. Sometimes another email serves no one. The continuous availability of the tool settles none of those questions.

Lu distinguishes three occasions for making that choice with AI. When a complicated job needs sorting out, he wants its speed. When he is trying to see what is in his head, he wants to write a few lines himself first, however rough they turn out. When he has already done enough analysis, he wants to have tea with the person concerned and actually talk. Each choice serves a different purpose. He wants to retain his reasons for choosing among them.

Moving from San Francisco to Taipei was partly a way of practicing that distinction. An empty calendar left him deciding each day what he wanted to do and wondering how to introduce himself apart from his job. A conversation, an unfamiliar street, or a song he kept replaying could remind him of something he cared about before it produced anything identifiable as work. Those experiences also gave him things to make. One use for the time a tool saves is to give the person in front of him his attention, without turning even that act of care into another score.

### A page for each day or a canvas for fragments

[16:19](https://www.youtube.com/watch?v=y9Sg--RDtks&t=979s)

Lu returns to the journal with two stripped-down possibilities. One gives each day a simple page, making it easy to revisit a particular date. The other resembles a scrapbook, an open canvas on which fragments from different times can sit together. Both answer the original brief. They differ in what someone can do when they come back to their memories.

Lu asks people to choose the version they would use, then find someone who picked the other one and ask why. Two people can agree that they want to record their lives and assume they understand each other. Put the two prototypes in front of them and they can discover that what each wants to hold on to is different. The versions can coexist, or people can try combining them. Choosing a winner would settle a product question while losing an opportunity to understand the other person.

Much of Lu’s own judgment developed through encounters like these. Someone tried something he made and found it odd. A colleague insisted on a detail he considered trivial. Someone else took apart an idea he felt sure about. Working through a misunderstanding could mean apologizing and trying again. Some of the relationships he treasures began that way. Those colleagues saw him confused, wrong, and exhausted, and they saw him genuinely excited about an idea. Over time he learned who noticed things he missed and who could take an idea apart without killing it.

### What a jazz ensemble gets from listening

[18:27](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1107s)

In Lu’s comparison, one musician carries the melody while another listens, leaves some space, and chooses when to come in. If the players lose the beat, they find it again by listening to one another. Counting who played the most notes would tell you little about how they managed it. The music can take a shape nobody planned and nobody could have produced alone.

An AI’s patience is genuinely helpful to Lu, and approaching it is often easier than approaching another person. What he fears losing, if he always chooses the easiest respondent, is the experience of getting in sync with someone different. A person who cares about an apparently trivial detail may change what he notices on the next project. That is a different contribution from helping him execute an intention he already has.

### One edit to the next polished answer

[19:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1185s)

Greater capability leaves the question of what Lu wants open. He sees no reason to wait for a grand mission before beginning. In the journal exercise, the order mattered. People saw a version, noticed something that felt wrong, considered a change, and learned something about their preference. Making can express a conviction someone already holds. It can also give someone their first occasion to discover one, perhaps by turning an idea they never dared show into a rough prototype another person can try.

Lu’s closing exercise begins the next time a polished answer arrives. Pause long enough to identify one decision in it that you have yet to think through. Pick the part you care about, make one small change, and, if possible, let someone else try it. For the journal, keeping a single reminder would suffice. The interesting questions are when you found out you wanted it and what the next person experiences when it appears. Lu still wants tools that bring making within reach of people who could scarcely participate before. Being able to make that edit, share it, and reconsider it with someone else is part of what participation means.

## Key takeaways

- Ask which feature you would remove from a generated design. Imagine using it the day after a missed entry to see what behavior its defaults encourage.
- Expose proposed edits before execution and let users adjust the result afterward. A proposed deletion can reveal a preference that a finished layout lets them test.
- Keep two plausible prototypes long enough to ask someone who chose the other one why. The difference can reveal a requirement the shared brief never expressed.

## Slides

### Stronger tools, greater unease

[0:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=51s) [A pale blue slide places the opening label at lower left and a large question about unease at upper right.](https://assets.ai.engineer/sites/aiecode2025/sha256/b36a47080eb98f8425cc549617fe7952ce2b1041c2815d916946abf4a3c248c0.webp)

Opening
STRONGER TOOLS
GREATER UNEASE

Our tools are stronger.
Why do we feel
more uneasy?

AIE

### 2,500 pull requests last month

[0:59](https://www.youtube.com/watch?v=y9Sg--RDtks&t=59s) [A pale speech bubble at upper left describes submitting 2,500 pull requests in a month. A small avatar appears beneath it.](https://assets.ai.engineer/sites/aiecode2025/sha256/8571387825fb7510a0ad7057ced960a23962cf8b94e9c6879d4f8a7f3cd97f7f.webp)

How I submitted 2,500
pull requests last month

AIE

### 40 PRs over drinks with friends

[1:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=73s) [A pale speech bubble at lower left describes submitting 40 PRs over drinks with friends. A yellow avatar appears beneath it.](https://assets.ai.engineer/sites/aiecode2025/sha256/044b9dba1c2d0b74890404bd0fd864e1fac2e4cc811149cbb34528ad1320eff9.webp)

I submitted 40 PRs
over drinks with friends

AIE

### Am I falling behind again?

[1:31](https://www.youtube.com/watch?v=y9Sg--RDtks&t=91s) [The slide shows “That’s amazing!” in dim gray above the brighter question “Am I falling behind again?”](https://assets.ai.engineer/sites/aiecode2025/sha256/dd4b8f5dc0d470cdba971e9c75de7cc399c1ce4d3d72c31fa8a3ca19bfa56768.webp)

That’s amazing!

Am I falling behind again?

AIE

### When is it enough?

[1:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=117s) [A two-line question in bold, pale lettering asks when it is enough if we can keep doing this forever.](https://assets.ai.engineer/sites/aiecode2025/sha256/3bfe15f573a4e6eff92a6747ddbb4cdaba35bc9091be08c19262e905a337bb5a.webp)

If we can keep doing this forever,
when is it enough?

AIE

### When we can do anything, do we still know what we want?

[2:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=165s) [A single question spans two left-aligned lines in pale lettering on a dark background.](https://assets.ai.engineer/sites/aiecode2025/sha256/c5bf1345d3b4a905acb3a91b4fbbdfbb0e173b34b4f3fd1cd8b226bb877b077b.webp)

When we can do anything,
do we still know what we want?

### Act I — Before Doing Became Proving

[3:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=183s) [A pale blue slide places the act heading at lower left and a larger sentence about making things at upper right.](https://assets.ai.engineer/sites/aiecode2025/sha256/3ff32e304ea47803a2e37736fbd4bb3b0c373309dbcf63fd787f6c88128e4cbc.webp)

Act I
BEFORE DOING
BECAME PROVING

Before making things
became a way to
prove ourselves.

### What’s special about me?

[3:31](https://www.youtube.com/watch?v=y9Sg--RDtks&t=211s) [A single question appears in two lines of bold white text on a dark background.](https://assets.ai.engineer/sites/aiecode2025/sha256/ae34ff5f2686b01a5b8d82144c518bd262dbf5b27509a499c90438b595cdc08e.webp)

If what I do is no longer special,
what’s special about me?

### Mac OS X desktop with overlapping windows

[3:58](https://www.youtube.com/watch?v=y9Sg--RDtks&t=238s) [A blue desktop shows a window of preference icons overlapping a news window with the Finder face. An Audio CD icon sits at upper right and the Dock runs along the bottom.](https://assets.ai.engineer/sites/aiecode2025/sha256/196ce548a90983931c4eeea135747850c256abb4821edbed0259411b53f3fcd0.webp)

Audio CD

Late-Breaking News
About Mac OS X

### Curiosity, play and enjoyment

[4:33](https://www.youtube.com/watch?v=y9Sg--RDtks&t=273s) [Three horizontal lines connect pairs of words in two columns, pairing curiosity with planning, play with output, and enjoyment with work.](https://assets.ai.engineer/sites/aiecode2025/sha256/0314ef7a8a1e53a0b1be7d4039c575717d68d5f090a0d9d1ea7a8d72eb0db738.webp)

Curiosity ─── Planning
Play      ─── Output
Enjoyment ─── Work

### What value does this add?

[4:46](https://www.youtube.com/watch?v=y9Sg--RDtks&t=286s) [Questions of different sizes fill the dark slide. The questions about value and shipping are largest. A partially legible question about retention appears at the top.](https://assets.ai.engineer/sites/aiecode2025/sha256/9d8d86966a50da680e1955b52487555fa41b215cf344bcea98c12bf65043e66b.webp)

retention?

What value does this add?

Have we collected the data?     Have we done the research?

When can we ship it?

Have we optimized the process?

### The right to enjoy making things

[5:12](https://www.youtube.com/watch?v=y9Sg--RDtks&t=312s) [A white sentence spans two lines on a dark background. The second line is larger and emphasizes the right to enjoy making things.](https://assets.ai.engineer/sites/aiecode2025/sha256/d1f96614466b32132968aa06d5af58496c335aa4434a5b3edb03c49cb54a521e.webp)

You don’t have to prove you’re better than anyone
to earn the right to enjoy making things.

### Act II — ANSWERED ≠ THOUGHT THROUGH

[5:46](https://www.youtube.com/watch?v=y9Sg--RDtks&t=346s) [On a pale blue slide, the Act II label sits at lower left. A larger statement about what a complete answer means appears on the right.](https://assets.ai.engineer/sites/aiecode2025/sha256/775a66e38d6f64d2fec3a641bbc1d3249850325d7c28ef08df271397d20a64c9.webp)

Act II
ANSWERED ≠
THOUGHT THROUGH

A complete answer
doesn’t mean I’ve
thought it through.

### Make me a tool to record my life.

[5:59](https://www.youtube.com/watch?v=y9Sg--RDtks&t=359s) [An orange chat bubble asks for a tool to record a life. A pale blue bubble replies, with a small orange face below.](https://assets.ai.engineer/sites/aiecode2025/sha256/fe723cceddb1229d43046c1b651a190f4759ca4cf4df56bfe7c89c38b9b872b3.webp)

Make me a tool to record my life.
Done. Take a look.

### Which journal feature would you drop?

[7:35](https://www.youtube.com/watch?v=y9Sg--RDtks&t=455s) [A show-of-hands question appears to the left of a dated journal entry. Four pale blue labels on the right list the features up for a vote.](https://assets.ai.engineer/sites/aiecode2025/sha256/55c3784fe6d7ce17d6a335bfcf061e017cabe55cf0acb44443d34dd9098aed36.webp)

Which feature
would you drop?
Hands up for each.

Journal
List   Free
Tue Oct 27
Wed Oct 28
Thu Oct 29
Fri Oct 30

Thu, October 29
Mood   ★★★★☆
Streak   6 days
Reminder   21:00

Played the same song again.
Not to remember the words.
Just to know it’s still there.

Weekly report
6 days writing · Avg. mood 3.8 · 6 days
A day itself doesn’t come with a score.

Mood score
Streak tracking
Daily reminders
Weekly report

### Streak, reminder and weekly report crossed out

[8:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=486s) [Three red X marks cross out the journal’s six-day streak, 21:00 reminder and weekly report. The selected date, four filled mood stars and three-line entry remain visible.](https://assets.ai.engineer/sites/aiecode2025/sha256/e74cdd1f5d3a541b463815735f8fcfc56bb8e3982ef20e4cd64a4ea45affddf5.webp)

Journal
List   Free
Tue Oct 27
Wed Oct 28
Thu Oct 29
Fri Oct 30
Sat Oct 31
Sun Nov 1
Mon Nov 2
Tue Nov 3
Wed Nov 4
Thu Nov 5
Fri Nov 6
Sat Nov 7
Sun Nov 8
Mon Nov 9

Thu, October 29
Mood   ★★★★☆
Streak   6 days
Reminder   21:00

Played the same song again.
Not to remember the words.
Just to know it’s still there.

Weekly report
6 days writing · Avg. mood

### Knowing what you want doesn’t have to come first

[9:07](https://www.youtube.com/watch?v=y9Sg--RDtks&t=547s) [Four lines of large white text on a dark background describe how a person’s wants can take shape after they start making something.](https://assets.ai.engineer/sites/aiecode2025/sha256/4359e149500f543fa879a0b48f727cdb87932d450ce414fb2a83acaf10d2977b.webp)

Knowing what you want
doesn’t have to come first.
Sometimes it takes shape
only after you start making.

### Removing the streak count and choosing journal features

[10:29](https://www.youtube.com/watch?v=y9Sg--RDtks&t=629s) [On the left, a user asks to remove a streak count and objects when the reply announces extra features. On the right, a journal preview offers four feature choices, with A highlighted beside Mood score.](https://assets.ai.engineer/sites/aiecode2025/sha256/f471e361c82ee28a98d74620d5282107e1e23c81014942dcdb5aba0035bcc006.webp)

Remove the streak count.
Take the pressure off.

Done! I added a bunch of
useful features, too.

I don’t want that.
Leave my design alone.

Build me a tool to
journal my life.

Searched 19 files and 4 sites.
Fine-tune in live preview.

Choose the features you want
A  Mood score
B  Streak tracking
C  Daily reminders
D  Weekly report

### React Effects Clean Up with a calendar preview

[10:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=651s) [A dark coding-agent workspace shows a task list beside a light calendar window. A green event bar appears in the calendar grid.](https://assets.ai.engineer/sites/aiecode2025/sha256/924d86e1fbde0b7678e40ed2c70cd8d016eef65c5ed2e74928c786308ad8261a.webp)

New Agent
React Effects Clean Up

research the codebase with subagents to find typical
react bugs and reduce number of effects / check if we
can do more with the app store

Initial Frontend Scaffolding
Create MissionControlView.tsx
Backend API Routes
Integrate APIs Together
Turn On Notifications

Review 4 Files    View Plan    Move to Cloud

### Reviewing code changes in React Effects Clean Up

[11:18](https://www.youtube.com/watch?v=y9Sg--RDtks&t=678s) [The task list appears on the left. On the right, a code diff marks removed lines in red and added lines in green, with a Create PR control above.](https://assets.ai.engineer/sites/aiecode2025/sha256/60bfb0cbe743d9e28adc05b29272ba9547ac1da7780f2c21347ce47b76cb4415.webp)

New Agent
React Effects Clean Up

research the codebase with subagents to find typical
react bugs and reduce number of effects / check if we
can do more with the app store

Initial Frontend Scaffolding
Create MissionControlView.tsx
Backend API Routes
Integrate APIs Together
Turn On Notifications

Review 4 Files    View Plan    Move to Cloud

Create PR
isTextStyled {

### A way in and a place to work

[11:50](https://www.youtube.com/watch?v=y9Sg--RDtks&t=710s) [Four lines of white text on a dark background pair wanting to understand with a way in, and wanting to make a change with a place to work.](https://assets.ai.engineer/sites/aiecode2025/sha256/f8f808af72aed75f3c76bc761416422a57aebc6f4361c88e2e09978159f3dd55.webp)

When I want to understand,
give me a way in.
When I want to change it,
give me a place to work.

### Visible assumptions, small changes, and taking back delegated work

[12:32](https://www.youtube.com/watch?v=y9Sg--RDtks&t=752s) [The number 1 appears above three instructions in white text about showing assumptions, trying small changes, and taking back delegated work.](https://assets.ai.engineer/sites/aiecode2025/sha256/70ed930eed863d7152328bf4f3a734b291a32c664cc29c1eac279e2c03349ee1.webp)

1

Make key assumptions visible.
Make each change small enough to try.
Let people take back delegated work
whenever they need to.

### More than accepting or rejecting the whole answer

[12:49](https://www.youtube.com/watch?v=y9Sg--RDtks&t=769s) [The number 2 appears above a four-line statement on giving people choices beyond confirming steps or accepting or rejecting an entire answer.](https://assets.ai.engineer/sites/aiecode2025/sha256/41d59df156cf606a4eb2ea9eee1d2d0f8db776110133975509c675f02a565881.webp)

2

Instead of confirming every step,
show people that they can
do more than accept or reject
the whole answer.

### Removing obstacles while preserving judgment

[13:00](https://www.youtube.com/watch?v=y9Sg--RDtks&t=780s) [The number 3 appears above a two-line statement about how good tools support people as they form judgments.](https://assets.ai.engineer/sites/aiecode2025/sha256/9e8fa8964efce12d0b4ac42b8f2822f499c75b7ee539199f7670c628c1c79a76.webp)

3

Good tools remove obstacles
while preserving how people form judgments.

### Act IV — Always There, Not Always Needed

[13:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=786s) [On a pale blue slide, the act heading sits at lower left. Larger text at upper right says that a tool within reach doesn’t have to be in constant use.](https://assets.ai.engineer/sites/aiecode2025/sha256/b9fa3f4007abae6dc91adddaf40a1c942ffbfe34a8037f00b92466fcc8938ae8.webp)

Act IV
ALWAYS THERE
NOT ALWAYS NEEDED

A tool within reach
doesn’t have to be
in constant use.

### Lamp-lit cutting mat and paper scraps

[13:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=793s) [A desk lamp lights a green cutting mat with drawing tools, a triangular ruler and white paper scraps. Sheets of paper and a roll of tape sit behind the mat.](https://assets.ai.engineer/sites/aiecode2025/sha256/ba2ad3213509f99f03a99ac8bf6a2f374b9ddb71fac8a1f31e4d553cfa37c417.webp)

### San Francisco and Taipei

[14:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=881s) [Side-by-side photos show a billboard with app icons in San Francisco and a street in Taipei. A horizontal line connects the city names.](https://assets.ai.engineer/sites/aiecode2025/sha256/5c938f05d63d84a004221a1bb28b3fb0c51d742db3f5f2a3e4dd57f0b45cf608.webp)

San Francisco ───── Taipei

### Life gives me things I want to make

[15:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=945s) [A single sentence in bold light lettering is centered on a dark background.](https://assets.ai.engineer/sites/aiecode2025/sha256/7c1d802f05c11a27b2d6db99988b90f2c10b164d3d52f8bb580043ea11d9fa33.webp)

Life gives me things I want to make.

### Act V — Stronger Together

[16:18](https://www.youtube.com/watch?v=y9Sg--RDtks&t=978s) [The act heading appears at lower left on a pale blue slide. At upper right, the phrase “we play together.” is in bold beneath two lines of regular type.](https://assets.ai.engineer/sites/aiecode2025/sha256/c340d70fd4c155f20806d5bde400332da40d639b1ec2ae5271b6a056fa593c2b.webp)

Act V
STRONGER
TOGETHER

Beyond growing
stronger alone,
we play together.

### Caring about the world

[16:27](https://www.youtube.com/watch?v=y9Sg--RDtks&t=987s) [Three left-aligned lines in bold light lettering appear against a dark background.](https://assets.ai.engineer/sites/aiecode2025/sha256/dbc32622566ba23aa3c8ab5af61e547465c4ed8df94af98b9e8d260b26b0143d.webp)

Caring about the world
means letting another person
into the way I think.

### A world that can change me

[19:29](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1169s) [Two lines of light text on a dark background describe wanting a world that can both respond to and change the speaker.](https://assets.ai.engineer/sites/aiecode2025/sha256/c26935f2fd4a05237b25a1e19c012d2a0376eda6c9e1d01096ea08eeb97e407a.webp)

I want more than a world that responds to me.
I also want a world that can change me.

### Begin before finding yourself

[19:44](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1184s) [On a pale blue slide, “Start making” appears in bold above two lines of serif text. A smaller ending label sits at the lower left.](https://assets.ai.engineer/sites/aiecode2025/sha256/092abc6854b006233f7d8eff650df5b2e3496e17e0d6b9092ce637ac270b2048.webp)

Start making
before you know
who you are.

Ending
BEGIN BEFORE
FINDING YOURSELF

### Making things helps you discover parts of yourself

[20:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1213s) [Four lines of bold, light text on a dark background describe how making things can reveal unfamiliar parts of yourself.](https://assets.ai.engineer/sites/aiecode2025/sha256/e1f8a62fd94a1c9eae554858e8aeeec91ecfcb996ff253b420ffd7aa3c5a7f2b.webp)

Making things does more
than express who you already are.
It helps you discover parts of yourself
you never knew.

### Together, we can make what we couldn’t make before

[22:49](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1369s) [Three left-aligned lines of light text on a dark background connect making things together with finding a way back to ourselves.](https://assets.ai.engineer/sites/aiecode2025/sha256/cc3eb3c838eb885c5bafe9ddf0e8c397edd51b3d202ccb1de54db48b323caa27.webp)

Together, we can make
what we couldn’t make before
and slowly find our way back to ourselves.

## Full transcript

[0:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=3s) Next up is Ryo. He was a founding engineer at Cursor.

[0:11](https://www.youtube.com/watch?v=y9Sg--RDtks&t=11s) Please welcome him. Designer, sorry, designer. Welcome, Ryo.

[0:16](https://www.youtube.com/watch?v=y9Sg--RDtks&t=16s) You'll want this one.

[0:19](https://www.youtube.com/watch?v=y9Sg--RDtks&t=19s) The handheld mic is more fun.

[0:25](https://www.youtube.com/watch?v=y9Sg--RDtks&t=25s) I'm a little nervous. Go easy on me.

[0:30](https://www.youtube.com/watch?v=y9Sg--RDtks&t=30s) OK.

[0:39](https://www.youtube.com/watch?v=y9Sg--RDtks&t=39s) Hi, I'm Ryo.

[0:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=41s) My talk today is called “When we can do anything,

[0:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=45s) do we still know what we want?” Let's get started.

[0:52](https://www.youtube.com/watch?v=y9Sg--RDtks&t=52s) To start: our tools are stronger. Why do we feel more uneasy?

[0:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=57s) I won't tell you how impressive AI is. You've seen it.

[1:00](https://www.youtube.com/watch?v=y9Sg--RDtks&t=60s) You've used it.

[1:02](https://www.youtube.com/watch?v=y9Sg--RDtks&t=62s) A website from one sentence. A feature in minutes.

[1:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=66s) You wake up to a whole night's work done for you,

[1:09](https://www.youtube.com/watch?v=y9Sg--RDtks&t=69s) including things you hadn't even thought of.

[1:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=74s) I was excited about that too. Things I could only imagine,

[1:18](https://www.youtube.com/watch?v=y9Sg--RDtks&t=78s) I could finally make.

[1:21](https://www.youtube.com/watch?v=y9Sg--RDtks&t=81s) Maybe you've felt something else too.

[1:25](https://www.youtube.com/watch?v=y9Sg--RDtks&t=85s) Someone makes something with a new model. Your first reaction is,

[1:28](https://www.youtube.com/watch?v=y9Sg--RDtks&t=88s) “Wow, that's amazing!”

[1:31](https://www.youtube.com/watch?v=y9Sg--RDtks&t=91s) Then comes the second one: “Am I falling behind again?”

[1:36](https://www.youtube.com/watch?v=y9Sg--RDtks&t=96s) Everyone seems smarter. Everyone's getting more done.

[1:40](https://www.youtube.com/watch?v=y9Sg--RDtks&t=100s) A skill you spent years developing now looks like a sentence

[1:44](https://www.youtube.com/watch?v=y9Sg--RDtks&t=104s) anyone can type into a box.

[1:48](https://www.youtube.com/watch?v=y9Sg--RDtks&t=108s) And you think, I should do more too. Learn a few more tools.

[1:53](https://www.youtube.com/watch?v=y9Sg--RDtks&t=113s) Spin up a few more agents.

[1:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=117s) But if we really can keep this up forever, when is it enough?

[2:05](https://www.youtube.com/watch?v=y9Sg--RDtks&t=125s) I've always believed in making tools.

[2:07](https://www.youtube.com/watch?v=y9Sg--RDtks&t=127s) I designed collaboration tools so people could build their own

[2:10](https://www.youtube.com/watch?v=y9Sg--RDtks&t=130s) systems.

[2:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=133s) Then I designed tools to help anyone make software.

[2:16](https://www.youtube.com/watch?v=y9Sg--RDtks&t=136s) I wanted to close the gap between imagination and reality.

[2:22](https://www.youtube.com/watch?v=y9Sg--RDtks&t=142s) I still believe in that. But as these tools let me do more,

[2:26](https://www.youtube.com/watch?v=y9Sg--RDtks&t=146s) something else matters to me.

[2:30](https://www.youtube.com/watch?v=y9Sg--RDtks&t=150s) Do they help me see what I want to make?

[2:33](https://www.youtube.com/watch?v=y9Sg--RDtks&t=153s) Or am I just trying to keep up?

[2:38](https://www.youtube.com/watch?v=y9Sg--RDtks&t=158s) I'm not here to debate whether to use AI,

[2:40](https://www.youtube.com/watch?v=y9Sg--RDtks&t=160s) or tell you that slow is always better.

[2:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=165s) When we can do anything, do we still know what we want?

[2:52](https://www.youtube.com/watch?v=y9Sg--RDtks&t=172s) And with tools this powerful,

[2:54](https://www.youtube.com/watch?v=y9Sg--RDtks&t=174s) how do we keep developing our own judgment,

[2:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=177s) figuring out what we care about, and living our own lives?

[3:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=183s) Part one. Before making things became proof of who we are.

[3:09](https://www.youtube.com/watch?v=y9Sg--RDtks&t=189s) A lot of that anxiety comes from a habit.

[3:12](https://www.youtube.com/watch?v=y9Sg--RDtks&t=192s) We use what we can do to prove who we are.

[3:18](https://www.youtube.com/watch?v=y9Sg--RDtks&t=198s) I'm a designer because I know design.

[3:20](https://www.youtube.com/watch?v=y9Sg--RDtks&t=200s) I'm an engineer because I can write code by hand.

[3:25](https://www.youtube.com/watch?v=y9Sg--RDtks&t=205s) So when a machine can do those things too,

[3:28](https://www.youtube.com/watch?v=y9Sg--RDtks&t=208s) my job is only part of the worry.

[3:30](https://www.youtube.com/watch?v=y9Sg--RDtks&t=210s) There's a more personal question.

[3:33](https://www.youtube.com/watch?v=y9Sg--RDtks&t=213s) If what I do isn't special anymore, what's special about me?

[3:39](https://www.youtube.com/watch?v=y9Sg--RDtks&t=219s) I don't have a neat answer.

[3:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=221s) But I remember how I felt about computers before any of this

[3:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=225s) became my job.

[3:49](https://www.youtube.com/watch?v=y9Sg--RDtks&t=229s) When I moved to Canada as a kid,

[3:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=231s) I saw that Apple computer with a screen that seemed to float.

[3:54](https://www.youtube.com/watch?v=y9Sg--RDtks&t=234s) How could something be that beautiful?

[3:58](https://www.youtube.com/watch?v=y9Sg--RDtks&t=238s) I couldn't afford one, so I made my own computer look like it.

[4:02](https://www.youtube.com/watch?v=y9Sg--RDtks&t=242s) Changed the icons and fonts.

[4:04](https://www.youtube.com/watch?v=y9Sg--RDtks&t=244s) Figured out how to make those buttons you wanted to lick.

[4:09](https://www.youtube.com/watch?v=y9Sg--RDtks&t=249s) I wasn't trying to become a designer. I liked it. I was curious.

[4:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=254s) I wanted a closer look, to see if I could make something feel

[4:17](https://www.youtube.com/watch?v=y9Sg--RDtks&t=257s) that way too.

[4:21](https://www.youtube.com/watch?v=y9Sg--RDtks&t=261s) Eventually that became my job. I was lucky.

[4:25](https://www.youtube.com/watch?v=y9Sg--RDtks&t=265s) I could make a living doing something I cared about and meet

[4:28](https://www.youtube.com/watch?v=y9Sg--RDtks&t=268s) people who cared about it too.

[4:33](https://www.youtube.com/watch?v=y9Sg--RDtks&t=273s) Then curiosity turned into planning. Play turned into output.

[4:37](https://www.youtube.com/watch?v=y9Sg--RDtks&t=277s) Taste turned into decisions I had to make on other people's

[4:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=281s) behalf.

[4:44](https://www.youtube.com/watch?v=y9Sg--RDtks&t=284s) You get good at answering:

[4:46](https://www.youtube.com/watch?v=y9Sg--RDtks&t=286s) What's it worth? When can we ship?

[4:49](https://www.youtube.com/watch?v=y9Sg--RDtks&t=289s) But do you remember why you wanted to make it

[4:53](https://www.youtube.com/watch?v=y9Sg--RDtks&t=293s) before anyone asked you to?

[4:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=297s) I don't take comfort in saying, “A machine could never do that.”

[5:01](https://www.youtube.com/watch?v=y9Sg--RDtks&t=301s) That line keeps moving.

[5:05](https://www.youtube.com/watch?v=y9Sg--RDtks&t=305s) What I want to hold on to is my own relationship with making things.

[5:11](https://www.youtube.com/watch?v=y9Sg--RDtks&t=311s) I shouldn't have to prove I'm better than anyone else

[5:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=314s) to keep doing what I love.

[5:18](https://www.youtube.com/watch?v=y9Sg--RDtks&t=318s) A machine can make ten versions in no time.

[5:21](https://www.youtube.com/watch?v=y9Sg--RDtks&t=321s) I can still stop at one detail and think, “That feels off.

[5:25](https://www.youtube.com/watch?v=y9Sg--RDtks&t=325s) I want to fix it myself.”

[5:28](https://www.youtube.com/watch?v=y9Sg--RDtks&t=328s) I can't always put that feeling into words right away.

[5:32](https://www.youtube.com/watch?v=y9Sg--RDtks&t=332s) But it's closer to what I really want to make

[5:34](https://www.youtube.com/watch?v=y9Sg--RDtks&t=334s) than thinking, “I should be doing more.”

[5:38](https://www.youtube.com/watch?v=y9Sg--RDtks&t=338s) So my question is, are these tools

[5:40](https://www.youtube.com/watch?v=y9Sg--RDtks&t=340s) skipping over that whole process too quickly?

[5:46](https://www.youtube.com/watch?v=y9Sg--RDtks&t=346s) Part two. A complete answer doesn't mean I've thought it through.

[5:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=351s) Say I ask,

[5:52](https://www.youtube.com/watch?v=y9Sg--RDtks&t=352s) “Make me a tool to record my day.”

[5:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=357s) Two minutes later, it says,

[5:59](https://www.youtube.com/watch?v=y9Sg--RDtks&t=359s) “Done. Take a look.”

[6:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=363s) Here's the journal it made.

[6:04](https://www.youtube.com/watch?v=y9Sg--RDtks&t=364s) It even explains why I need each feature.

[6:08](https://www.youtube.com/watch?v=y9Sg--RDtks&t=368s) Mood scores help me spot trends.

[6:11](https://www.youtube.com/watch?v=y9Sg--RDtks&t=371s) Daily reminders help me build a habit.

[6:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=374s) Weekly reports help me reflect.

[6:17](https://www.youtube.com/watch?v=y9Sg--RDtks&t=377s) Sounds reasonable. I'd probably say, “Sure. Let's go with that.”

[6:23](https://www.youtube.com/watch?v=y9Sg--RDtks&t=383s) But hang on. Have I really thought about why my life needs a

[6:27](https://www.youtube.com/watch?v=y9Sg--RDtks&t=387s) score?

[6:30](https://www.youtube.com/watch?v=y9Sg--RDtks&t=390s) Did I decide that keeping a journal means writing every day?

[6:35](https://www.youtube.com/watch?v=y9Sg--RDtks&t=395s) Or was its explanation of what others do so complete that I

[6:39](https://www.youtube.com/watch?v=y9Sg--RDtks&t=399s) mistook it for my own understanding?

[6:44](https://www.youtube.com/watch?v=y9Sg--RDtks&t=404s) That's what worries me. Sometimes I don't form an opinion of my

[6:47](https://www.youtube.com/watch?v=y9Sg--RDtks&t=407s) own before I ask the tool for help.

[6:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=411s) I see what it says first. I can't think of an objection,

[6:54](https://www.youtube.com/watch?v=y9Sg--RDtks&t=414s) so I go with it.

[6:58](https://www.youtube.com/watch?v=y9Sg--RDtks&t=418s) The answer might be right.

[6:59](https://www.youtube.com/watch?v=y9Sg--RDtks&t=419s) It might even be better than anything I'd imagined.

[7:04](https://www.youtube.com/watch?v=y9Sg--RDtks&t=424s) But if I do that every time,

[7:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=426s) I may get good at judging whether something sounds reasonable,

[7:10](https://www.youtube.com/watch?v=y9Sg--RDtks&t=430s) while getting less and less practice thinking a problem through

[7:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=433s) from my own experience.

[7:16](https://www.youtube.com/watch?v=y9Sg--RDtks&t=436s) Having an answer doesn't mean I've finished thinking.

[7:21](https://www.youtube.com/watch?v=y9Sg--RDtks&t=441s) An obvious mistake is easy to spot.

[7:23](https://www.youtube.com/watch?v=y9Sg--RDtks&t=443s) What's harder to see are the questions a complete answer covers

[7:27](https://www.youtube.com/watch?v=y9Sg--RDtks&t=447s) up before we've even had a chance to ask them.

[7:32](https://www.youtube.com/watch?v=y9Sg--RDtks&t=452s) Let's try something. Look at this journal. For a moment,

[7:36](https://www.youtube.com/watch?v=y9Sg--RDtks&t=456s) set aside what features it might be missing.

[7:40](https://www.youtube.com/watch?v=y9Sg--RDtks&t=460s) Mood scores, writing streaks, daily reminders, weekly reports.

[7:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=465s) Which one would you take out?

[7:52](https://www.youtube.com/watch?v=y9Sg--RDtks&t=472s) Hands up if you'd drop mood scores.

[7:56](https://www.youtube.com/watch?v=y9Sg--RDtks&t=476s) Writing streaks?

[8:00](https://www.youtube.com/watch?v=y9Sg--RDtks&t=480s) Daily reminders?

[8:04](https://www.youtube.com/watch?v=y9Sg--RDtks&t=484s) And weekly reports?

[8:07](https://www.youtube.com/watch?v=y9Sg--RDtks&t=487s) I'd probably drop the streaks, the daily reminders,

[8:10](https://www.youtube.com/watch?v=y9Sg--RDtks&t=490s) and the weekly reports.

[8:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=493s) Miss one day, and I'd feel like I owed somebody.

[8:18](https://www.youtube.com/watch?v=y9Sg--RDtks&t=498s) I want to record my life. I don't want homework.

[8:23](https://www.youtube.com/watch?v=y9Sg--RDtks&t=503s) Maybe you're the opposite.

[8:24](https://www.youtube.com/watch?v=y9Sg--RDtks&t=504s) That reminder is exactly what you need.

[8:26](https://www.youtube.com/watch?v=y9Sg--RDtks&t=506s) Maybe keeping a streak matters to you.

[8:31](https://www.youtube.com/watch?v=y9Sg--RDtks&t=511s) Or maybe it's all wrong.

[8:32](https://www.youtube.com/watch?v=y9Sg--RDtks&t=512s) You want an infinite canvas to arrange memories and bits of your

[8:34](https://www.youtube.com/watch?v=y9Sg--RDtks&t=514s) life however you like.

[8:39](https://www.youtube.com/watch?v=y9Sg--RDtks&t=519s) No right answer here.

[8:43](https://www.youtube.com/watch?v=y9Sg--RDtks&t=523s) All we did was pause. But something happened in that pause.

[8:47](https://www.youtube.com/watch?v=y9Sg--RDtks&t=527s) We picked out one decision, noticed we felt something about it,

[8:52](https://www.youtube.com/watch?v=y9Sg--RDtks&t=532s) and started to say why.

[8:55](https://www.youtube.com/watch?v=y9Sg--RDtks&t=535s) That vague idea, “I want to record my life,”

[8:58](https://www.youtube.com/watch?v=y9Sg--RDtks&t=538s) finally started taking on a shape of its own.

[9:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=543s) You don't always have to know what you want before you start.

[9:08](https://www.youtube.com/watch?v=y9Sg--RDtks&t=548s) Sometimes you find out by making it.

[9:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=554s) Part three. A service counter and a workbench.

[9:19](https://www.youtube.com/watch?v=y9Sg--RDtks&t=559s) So how can our tools help?

[9:23](https://www.youtube.com/watch?v=y9Sg--RDtks&t=563s) I'd hate to say you have to puzzle through everything first or

[9:27](https://www.youtube.com/watch?v=y9Sg--RDtks&t=567s) you're being lazy.

[9:30](https://www.youtube.com/watch?v=y9Sg--RDtks&t=570s) Some jobs I genuinely want to hand off. Move the data.

[9:33](https://www.youtube.com/watch?v=y9Sg--RDtks&t=573s) Convert the file. I don't need a lesson every time.

[9:38](https://www.youtube.com/watch?v=y9Sg--RDtks&t=578s) Some things I'm still figuring out.

[9:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=581s) I want a tool that lets me keep exploring,

[9:43](https://www.youtube.com/watch?v=y9Sg--RDtks&t=583s) rather than one that closes the question for me.

[9:49](https://www.youtube.com/watch?v=y9Sg--RDtks&t=589s) Compare these two screens.

[9:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=591s) The one on the left is like a service counter.

[9:56](https://www.youtube.com/watch?v=y9Sg--RDtks&t=596s) It processes my request and hands me a new version.

[9:59](https://www.youtube.com/watch?v=y9Sg--RDtks&t=599s) If it happens to be good, great. I'll use it.

[10:04](https://www.youtube.com/watch?v=y9Sg--RDtks&t=604s) But if it removes a reminder I like or changes the design,

[10:07](https://www.youtube.com/watch?v=y9Sg--RDtks&t=607s) I have to explain again what to keep and what to drop.

[10:12](https://www.youtube.com/watch?v=y9Sg--RDtks&t=612s) If all I see is the finished product,

[10:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=614s) my choices boil down to accept, reject, or explain it again.

[10:21](https://www.youtube.com/watch?v=y9Sg--RDtks&t=621s) The one on the right is more like a workbench. Same request,

[10:24](https://www.youtube.com/watch?v=y9Sg--RDtks&t=624s) but I can see what it's planning to change,

[10:25](https://www.youtube.com/watch?v=y9Sg--RDtks&t=625s) which features it'll remove,

[10:32](https://www.youtube.com/watch?v=y9Sg--RDtks&t=632s) or what it wants to suggest.

[10:36](https://www.youtube.com/watch?v=y9Sg--RDtks&t=636s) Seeing that makes me think: wait.

[10:38](https://www.youtube.com/watch?v=y9Sg--RDtks&t=638s) Maybe I want to keep that reminder.

[10:42](https://www.youtube.com/watch?v=y9Sg--RDtks&t=642s) I can step into the plan before it starts. Once it's done,

[10:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=645s) I can fine-tune the result in a live preview.

[10:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=651s) There's more to the difference than a few extra buttons. Here,

[10:55](https://www.youtube.com/watch?v=y9Sg--RDtks&t=655s) I can put my hands on the thing I'm changing.

[10:59](https://www.youtube.com/watch?v=y9Sg--RDtks&t=659s) I can step inside instead of waiting for an answer.

[11:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=663s) Sometimes I only realize I want that reminder when I see it's

[11:07](https://www.youtube.com/watch?v=y9Sg--RDtks&t=667s) about to disappear.

[11:09](https://www.youtube.com/watch?v=y9Sg--RDtks&t=669s) Sometimes I try my own design and realize the layout won't work.

[11:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=674s) Or I can think it through before I run the plan.

[11:20](https://www.youtube.com/watch?v=y9Sg--RDtks&t=680s) The tool helps me discover what I think.

[11:23](https://www.youtube.com/watch?v=y9Sg--RDtks&t=683s) I don't have to explain the idea perfectly at the start.

[11:27](https://www.youtube.com/watch?v=y9Sg--RDtks&t=687s) That's why, when I design software, I care about files, plans,

[11:33](https://www.youtube.com/watch?v=y9Sg--RDtks&t=693s) edits, what changed between versions,

[11:35](https://www.youtube.com/watch?v=y9Sg--RDtks&t=695s) and where a person can step in.

[11:39](https://www.youtube.com/watch?v=y9Sg--RDtks&t=699s) Nobody should have to read every line of code.

[11:42](https://www.youtube.com/watch?v=y9Sg--RDtks&t=702s) Showing every detail isn't the only way to be transparent.

[11:47](https://www.youtube.com/watch?v=y9Sg--RDtks&t=707s) What matters is that when I want to understand something,

[11:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=711s) there's a way in. When I want to change it, I can.

[11:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=717s) Chat can be part of a workbench too.

[11:59](https://www.youtube.com/watch?v=y9Sg--RDtks&t=719s) I wouldn't put one interface or one brand permanently on either

[12:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=723s) side.

[12:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=726s) The question is: for this particular task,

[12:08](https://www.youtube.com/watch?v=y9Sg--RDtks&t=728s) can I change just one part?

[12:11](https://www.youtube.com/watch?v=y9Sg--RDtks&t=731s) Can I see why a decision was made? Can I take back control?

[12:16](https://www.youtube.com/watch?v=y9Sg--RDtks&t=736s) Does the interface give me room to think and explore?

[12:22](https://www.youtube.com/watch?v=y9Sg--RDtks&t=742s) I want tools that involve me more.

[12:24](https://www.youtube.com/watch?v=y9Sg--RDtks&t=744s) I don't need proof they can do without me.

[12:29](https://www.youtube.com/watch?v=y9Sg--RDtks&t=749s) If you build tools too, here are three small places to start.

[12:35](https://www.youtube.com/watch?v=y9Sg--RDtks&t=755s) First, make important assumptions visible.

[12:38](https://www.youtube.com/watch?v=y9Sg--RDtks&t=758s) Make each change small enough to try.

[12:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=761s) Let people take back a task they've handed off.

[12:47](https://www.youtube.com/watch?v=y9Sg--RDtks&t=767s) Second, you don't need a confirmation button at every step.

[12:50](https://www.youtube.com/watch?v=y9Sg--RDtks&t=770s) Show people they have more choices than accepting or rejecting

[12:53](https://www.youtube.com/watch?v=y9Sg--RDtks&t=773s) the whole answer.

[12:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=777s) Third, remove the obstacles.

[13:00](https://www.youtube.com/watch?v=y9Sg--RDtks&t=780s) Leave people the chance to form their own judgment along the way.

[13:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=786s) Part four. It's always there. Do I always need it?

[13:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=793s) Even at a great workbench,

[13:15](https://www.youtube.com/watch?v=y9Sg--RDtks&t=795s) you don't have to spend your life sitting there.

[13:20](https://www.youtube.com/watch?v=y9Sg--RDtks&t=800s) We're still figuring out what kind of relationship we want with

[13:23](https://www.youtube.com/watch?v=y9Sg--RDtks&t=803s) AI.

[13:25](https://www.youtube.com/watch?v=y9Sg--RDtks&t=805s) It can help me think. Or I can stop thinking for myself.

[13:30](https://www.youtube.com/watch?v=y9Sg--RDtks&t=810s) It can save me time, then give me more reasons to fill it right

[13:34](https://www.youtube.com/watch?v=y9Sg--RDtks&t=814s) back up.

[13:37](https://www.youtube.com/watch?v=y9Sg--RDtks&t=817s) A better tool still can't tell me how to use it.

[13:43](https://www.youtube.com/watch?v=y9Sg--RDtks&t=823s) Take email. Having it doesn't mean there's an email to send about

[13:46](https://www.youtube.com/watch?v=y9Sg--RDtks&t=826s) everything, any time of day.

[13:49](https://www.youtube.com/watch?v=y9Sg--RDtks&t=829s) Some things are worth writing down. Some are better face to face.

[13:53](https://www.youtube.com/watch?v=y9Sg--RDtks&t=833s) Some can wait for a reply.

[13:56](https://www.youtube.com/watch?v=y9Sg--RDtks&t=836s) Sometimes nobody needs one more email.

[14:01](https://www.youtube.com/watch?v=y9Sg--RDtks&t=841s) The tool is always there.

[14:02](https://www.youtube.com/watch?v=y9Sg--RDtks&t=842s) When to use it and what to use it for are still our choices.

[14:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=846s) AI is the same.

[14:10](https://www.youtube.com/watch?v=y9Sg--RDtks&t=850s) Sometimes I want speed. Help me sort through a complicated job.

[14:15](https://www.youtube.com/watch?v=y9Sg--RDtks&t=855s) Sometimes I want to write a few lines myself first.

[14:19](https://www.youtube.com/watch?v=y9Sg--RDtks&t=859s) They might be terrible. I want to see what's in my head.

[14:24](https://www.youtube.com/watch?v=y9Sg--RDtks&t=864s) Sometimes I've had enough analysis.

[14:26](https://www.youtube.com/watch?v=y9Sg--RDtks&t=866s) What I need is to go have tea with that person and actually talk.

[14:32](https://www.youtube.com/watch?v=y9Sg--RDtks&t=872s) None of these choices is more virtuous.

[14:35](https://www.youtube.com/watch?v=y9Sg--RDtks&t=875s) I want to keep my reasons for choosing differently.

[14:39](https://www.youtube.com/watch?v=y9Sg--RDtks&t=879s) Leaving San Francisco and moving to Taipei was partly a way of

[14:43](https://www.youtube.com/watch?v=y9Sg--RDtks&t=883s) practicing that.

[14:46](https://www.youtube.com/watch?v=y9Sg--RDtks&t=886s) I still wanted to make things.

[14:47](https://www.youtube.com/watch?v=y9Sg--RDtks&t=887s) I just didn't want to explain everything in my life in terms of

[14:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=891s) whether it helped my work.

[14:56](https://www.youtube.com/watch?v=y9Sg--RDtks&t=896s) Stepping out of my old routine didn't immediately make me sure of

[14:59](https://www.youtube.com/watch?v=y9Sg--RDtks&t=899s) myself.

[15:01](https://www.youtube.com/watch?v=y9Sg--RDtks&t=901s) With an empty calendar, I had to ask myself every day:

[15:05](https://www.youtube.com/watch?v=y9Sg--RDtks&t=905s) what do I want to do?

[15:08](https://www.youtube.com/watch?v=y9Sg--RDtks&t=908s) If I can't introduce myself by my job, who am I?

[15:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=913s) I started to feel that time could be worthwhile even if it

[15:16](https://www.youtube.com/watch?v=y9Sg--RDtks&t=916s) produced nothing right away.

[15:19](https://www.youtube.com/watch?v=y9Sg--RDtks&t=919s) Talking to another person. A street I've never walked down.

[15:23](https://www.youtube.com/watch?v=y9Sg--RDtks&t=923s) A song I keep replaying.

[15:26](https://www.youtube.com/watch?v=y9Sg--RDtks&t=926s) Any of them can remind me I still care about something.

[15:30](https://www.youtube.com/watch?v=y9Sg--RDtks&t=930s) That feeling is mine.

[15:35](https://www.youtube.com/watch?v=y9Sg--RDtks&t=935s) Life and creative work aren't competing.

[15:37](https://www.youtube.com/watch?v=y9Sg--RDtks&t=937s) Squeezing life out won't leave you with more creativity.

[15:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=941s) Life gives me things I want to make.

[15:46](https://www.youtube.com/watch?v=y9Sg--RDtks&t=946s) So what do I hope these tools give us in the end?

[15:50](https://www.youtube.com/watch?v=y9Sg--RDtks&t=950s) More than maximizing output,

[15:53](https://www.youtube.com/watch?v=y9Sg--RDtks&t=953s) I want more time to care for other people and the world that we

[15:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=957s) live in.

[16:01](https://www.youtube.com/watch?v=y9Sg--RDtks&t=961s) I don't want to turn caring into another score to maximize.

[16:05](https://www.youtube.com/watch?v=y9Sg--RDtks&t=965s) I mean giving time and attention to the person in front of you.

[16:10](https://www.youtube.com/watch?v=y9Sg--RDtks&t=970s) Having room to understand something you haven't figured out yet.

[16:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=973s) Being able to do something about what you care about.

[16:19](https://www.youtube.com/watch?v=y9Sg--RDtks&t=979s) Part five. Beyond getting better on our own: playing together.

[16:24](https://www.youtube.com/watch?v=y9Sg--RDtks&t=984s) Caring about the world also means letting someone else into my

[16:28](https://www.youtube.com/watch?v=y9Sg--RDtks&t=988s) thinking.

[16:31](https://www.youtube.com/watch?v=y9Sg--RDtks&t=991s) Back to our journal. We changed one decision.

[16:34](https://www.youtube.com/watch?v=y9Sg--RDtks&t=994s) It's closer to the journal I want now,

[16:36](https://www.youtube.com/watch?v=y9Sg--RDtks&t=996s) but why would it be the one you want?

[16:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1001s) Here's a simple page for each day,

[16:43](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1003s) with all the extras stripped out. Or it could be a scrapbook:

[16:47](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1007s) an open canvas with little fragments of memory.

[16:52](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1012s) Which of these two would you rather use?

[16:56](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1016s) Daily pages make a particular day easy to revisit.

[17:00](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1020s) A freeform canvas brings memories from different times together.

[17:05](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1025s) We don't have to pick a winner. We can keep both versions.

[17:09](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1029s) We could even put them together.

[17:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1033s) Find someone who picked the other one. Ask them why.

[17:18](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1038s) Now we have more than a finished product.

[17:20](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1040s) We have something that helps us understand each other.

[17:26](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1046s) We both say, “I want to record my life,”

[17:29](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1049s) and think we mean the same thing. Then we see these two versions,

[17:33](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1053s) and realize we're trying to hold on to different things.

[17:39](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1059s) That's how a lot of my judgment at work took shape.

[17:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1061s) Sitting alone didn't get me there.

[17:43](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1063s) Someone tried what I'd made and said it felt odd.

[17:47](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1067s) Someone insists on a detail I think is trivial.

[17:50](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1070s) Someone takes apart an idea I was sure about.

[17:53](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1073s) That's how I see what I missed.

[17:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1077s) That can be uncomfortable.

[17:58](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1078s) Working together means misunderstandings and disagreements.

[18:02](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1082s) You apologize and try again.

[18:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1086s) It's also where a lot of relationships I treasure began.

[18:11](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1091s) People have seen me confused, wrong, exhausted.

[18:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1093s) They've also seen me genuinely excited.

[18:17](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1097s) Over time you learn who spots the things you miss,

[18:20](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1100s) and who can take an idea apart without killing it.

[18:27](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1107s) It's like a jazz band. Nobody's trying to play the most notes.

[18:31](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1111s) Nobody needs to prove they're the best player.

[18:35](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1115s) One person carries the melody. Somebody else listens,

[18:38](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1118s) leaves a little space, then comes in.

[18:42](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1122s) Lose the beat? You listen to each other and find it again.

[18:47](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1127s) What comes out is something none of you had in mind beforehand,

[18:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1131s) and none of you could have made alone.

[18:55](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1135s) AI is convenient. It's patient.

[18:58](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1138s) Often it's easier than talking to another person.

[19:02](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1142s) Sometimes that really does help me. I want to acknowledge that.

[19:07](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1147s) If I always choose whoever responds to me most easily,

[19:10](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1150s) I worry about what I'm missing:

[19:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1153s) the chance to get in sync with someone different from me.

[19:18](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1158s) I hope our tools don't lock each of us inside a more efficient

[19:22](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1162s) little world.

[19:25](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1165s) I don't need a world that just responds better to me.

[19:28](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1168s) I want a world that can change me.

[19:33](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1173s) Tools let me do things I couldn't do alone.

[19:37](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1177s) Other people make things possible that I never would have thought

[19:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1181s) of.

[19:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1185s) The final act. You can start making things before you know who

[19:48](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1188s) you are.

[19:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1191s) So back to today's question. When we can do anything,

[19:54](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1194s) do we still know what we want? My answer is still:

[19:58](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1198s) not necessarily.

[20:01](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1201s) But I don't think that means there's something missing in us.

[20:05](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1205s) And we don't need to find some grand mission before we can begin.

[20:10](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1210s) Sometimes finding yourself has little to do with suddenly knowing

[20:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1214s) how to introduce yourself.

[20:18](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1218s) It happens in little choices. You find yourself thinking:

[20:21](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1221s) this is what I care about.

[20:24](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1224s) This part I won't take as it comes. That part?

[20:27](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1227s) I'd like to spend more time on it.

[20:31](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1231s) Like just now. We didn't write a product spec first.

[20:34](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1234s) We saw a version, felt something was off, changed it,

[20:37](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1237s) and found out what we wanted.

[20:41](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1241s) Making things gives you a way to express who you are.

[20:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1245s) It also lets you meet parts of yourself you didn't know.

[20:52](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1252s) To me, finding yourself and going beyond yourself go together.

[20:57](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1257s) Making things shows you what you care about.

[21:00](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1260s) And with tools and other people,

[21:02](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1262s) you can reach places you couldn't reach before.

[21:06](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1266s) That doesn't mean outproducing yesterday's self every day,

[21:09](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1269s) or turning yourself into a product that always needs an upgrade.

[21:14](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1274s) Maybe you can now change something you used to have to accept.

[21:20](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1280s) An idea you never dared show anyone can now be a rough prototype

[21:24](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1284s) you give someone to try.

[21:27](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1287s) Maybe you can finally do that one thing you always wanted to do

[21:31](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1291s) and never thought you could.

[21:35](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1295s) If you take one thing away today,

[21:37](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1297s) I hope it goes beyond trying one more tool.

[21:40](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1300s) Next time you see a polished answer, give yourself a moment. Ask:

[21:45](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1305s) which decision in here haven't I thought through for myself?

[21:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1311s) Pick a part you care about. Try one small change.

[21:54](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1314s) If you get a chance, show it to someone else.

[22:00](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1320s) You needn't change everything or prove you're better than the

[22:03](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1323s) machine.

[22:05](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1325s) As long as you bring what you feel, what you think,

[22:08](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1328s) and what you care about to that process,

[22:13](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1333s) you're doing more than standing by and approving a result.

[22:16](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1336s) You're shaping your own life.

[22:19](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1339s) I still believe in tools.

[22:20](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1340s) I hope they give people who couldn't participate before a chance

[22:23](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1343s) to express themselves.

[22:26](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1346s) I hope we can make things once beyond us.

[22:29](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1349s) And we shouldn't have to hand over our whole lives just to keep

[22:33](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1353s) the system running.

[22:36](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1356s) I don't want a tool that lives my life for me.

[22:39](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1359s) I want one that helps me feel things more clearly,

[22:43](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1363s) explore freely, and care for the world around me,

[22:48](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1368s) make things with others that we couldn't make before,

[22:51](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1371s) and slowly find my way back to myself.

[22:56](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1376s) Thank you.

[22:59](https://www.youtube.com/watch?v=y9Sg--RDtks&t=1379s) Thank you, everyone.
