Thariq Shihipar
Worked on Claude Code at Anthropic for more than a year, until October 2026, building demos and prototypes of how to use Claude well; previously YC W20, South Park Commons, MIT Media Lab.
Thariq Shihipar did not write this page. ppll compiled it from their own words, each linked to the source. Is this you? · Report a mistake
Latest x.com
What they believe
-
Making your own software is hard even with AI, because it is hard to know what you really want.
making your own software is tough, even with AI, because it's hard to know what you really want I didn't end up sticking with this, the modes weren't the right way for me to organize my work gonna a different take that’s more to-do oriented
x.comSaid 18 Nov 2026This on korrents
-
Software factories are a temporary problem: the models will learn to build them, and the lasting question is a workspace where agents collaborate with you.
software factories feel like a temporary problem. the models will learn to build them for you the more interesting question to me is how to build a workspace where agents can collaborate with you
x.comSaid 16 Nov 2026This on korrents
The bold line is korrents’ wording; the quote is theirs. These 2: what they wrote before what they said aloud, one per source, newest first.
Feed
As its own page →Hiding
21 November
20 November
19 November
18 November
-
korrents.com
Making your own software is hard even with AI, because it is hard to know what you really want.Their words
making your own software is tough, even with AI, because it's hard to know what you really want I didn't end up sticking with this, the modes weren't the right way for me to organize my work gonna a different take that’s more to-do oriented
16 November
-
Their words
software factories feel like a temporary problem. the models will learn to build them for you the more interesting question to me is how to build a workspace where agents can collaborate with you
13 November
11 November
10 November
-
korrents.com
AI tools should show how to collaborate with models on a vision rather than chase one-shot demos.Their words
we need fewer one shot demos and more tools that show how you can collaborate with models to bring your vision to life
7 November
6 November
5 November
Nothing matches.
Beliefs
Korrents What they believe 10 beliefs — each backed by an exact quote.
Each is a — compiled by korrents.com, not by them: the one-line wordings are korrents', the quotes are theirs.
Recent
Making your own software is hard even with AI, because it is hard to know what you really want.
making your own software is tough, even with AI, because it's hard to know what you really want I didn't end up sticking with this, the modes weren't the right way for me to organize my work gonna a different take that’s more to-do oriented
@trq212 on X Said 18 Nov 2026
Software factories are a temporary problem: the models will learn to build them, and the lasting question is a workspace where agents collaborate with you.
software factories feel like a temporary problem. the models will learn to build them for you the more interesting question to me is how to build a workspace where agents can collaborate with you
@trq212 on X Said 16 Nov 2026
AI tools should show how to collaborate with models on a vision rather than chase one-shot demos.
we need fewer one shot demos and more tools that show how you can collaborate with models to bring your vision to life
@trq212 on X Said 10 Nov 2026
Show 7 more
The smartest models will need fewer tokens because they will not need to check their work; they will just be right.
this is basically the bitter lesson at work the smartest models will need fewer and fewer tokens, they won't need to verify their work or check it for bugs, they will just be right
@trq212 on X Said 3 Nov 2026
You do not need unlimited tokens to do great work with AI, and intelligence will never be more expensive than it is now.
one week in my takeaway is that you don't need unlimited tokens to do great work, the $200 plans are amazing and this is the most expensive intelligence will ever be
@trq212 on X Said 2 Nov 2026
The answer to economic doomerism about AI is to believe in AGI even more: a well-aligned superintelligence that explains its work and makes each person more effective.
the solution to economic doomerism is to be even more AGI-pilled I believe in a well-aligned ASI that can explain its work and bring us along, one that makes each individual maximally effective
@trq212 on X Said 30 Oct 2026
Coding agents can induce a mania of feeling busy while nothing is really happening, often as a way of avoiding something else.
There’s 100% a kind of mania you can get into with coding agents, where it feels like you’re getting a lot done and working all the time but nothing is really happening. I think it's often a form of avoidance of something else in life.
@trq212 on X Said 29 Oct 2026
Most people multitasking with coding agents would do better to slow down and give their top priority their full attention.
I think most people multi-tasking with agents could benefit from slowing down most of the stuff you're doing wont really matter- what's your P0? what would it look like if you gave it your full attention?
@trq212 on X Said 19 Oct 2026
We should not settle for a superintelligence that is opaque; AI should build understanding with people instead of cutting them out.
I think Terrence Tao is directionally correct, we shouldn’t settle for a super intelligence that is opaque. A key signal of intelligence is communication. We should strive to make products and models that build understanding with humans instead of disintermediating them.
@trq212 on X Said 11 Oct 2026
The right response to more capable models is not more shipped features; it is spending that capability on understanding users and prototyping, so what ships actually works.
the right way to use model capabilities is not to ship 10x more features to prod it's to spend more time understanding your users, trying experiments, building prototypes, learning about things you don't understand so that you can ship things that actually work
Thariq (@trq212) on X Said 23 Sept 2026
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
Where they publish
x.com @trq212 Notes on building with model capabilities, prototyping and Claude Code.
Short posts on what he's building at Anthropic and how he thinks about using model capabilities well, rather than just shipping more features.
Recent
- when I see some of the greats using models to make huge advances, like @__alpoge__ on the Riemann hypothesis or @jarredsumner on ahead-of-time javascript I feel like I'm seeing cyborgs. humans and machines working in beautiful tandem, pushing the frontier together. 21 Nov 2026
- I think I’ve figured out the gameplay loop! It’s a sports anime game, but the sport is elemental bending. You fight in tournaments alongside your squad, and between matches they have their own stories you can get into, like in Mass Effect or Dragon Age. 20 Nov 2026
- I asked Claude how I could save space on my computer 20 Nov 2026
Show 17 more
- Software is both art and utility. AI is both an Industrial Revolution and a Renaissance. Factories are for utilities, at AIE Code I talked about how we can make better art. 19 Nov 2026
- an age of miracles 19 Nov 2026
- making your own software is tough, even with AI, because it's hard to know what you really want I didn't end up sticking with this, the modes weren't the right way for me to organize my work gonna a different take that’s more to-do oriented 18 Nov 2026
- software factories feel like a temporary problem. the models will learn to build them for you the more interesting question to me is how to build a workspace where agents can collaborate with you 16 Nov 2026
- a new arena and two new characters 16 Nov 2026
- it’s time to retvrn 13 Nov 2026
- Swyx gave me the “closing keynote” of AIE code, aka everyone is tired or flying out so if you’re not one of those please come to mine tomorrow at 5pm it’s called “the sweet side of the bitter lesson” 11 Nov 2026
- we need fewer one shot demos and more tools that show how you can collaborate with models to bring your vision to life 10 Nov 2026
- too many people building agents and not enough people using software to improve the quality of fish 7 Nov 2026
- a pattern I’m liking: creator apps like Instagram Edits and Roblox Studio shipping with built-in MCPs, so agents can do some of the work while power users can collaborate and continue the work in tools that they're used to 6 Nov 2026
- feels like it's really coming together 6 Nov 2026
- one of my favorite prompts for agents like Instinct is location based reminders e.g. "Keep track of when I've done my laundry. If it's been more than 5 days, remind me 30 minutes after I get home on a week day and bug me until I confirm or delay" 5 Nov 2026
- manifesting ⠀⠀⠀⠀ 🕯️🕯️🕯️🕯️🕯️ ⠀⠀🕯️⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀🕯️ 🕯️⠀⠀Apple lets apps⠀ 🕯️ 🕯️⠀⠀⠀build Shortcuts 🕯️ 🕯️⠀⠀⠀⠀. for you⠀⠀ ⠀ 🕯️ ⠀⠀🕯️⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀🕯️ ⠀⠀⠀⠀ 🕯️🕯️🕯️🕯️🕯️ 4 Nov 2026
- this is basically the bitter lesson at work the smartest models will need fewer and fewer tokens, they won't need to verify their work or check it for bugs, they will just be right 3 Nov 2026
- one week in my takeaway is that you don't need unlimited tokens to do great work, the $200 plans are amazing and this is the most expensive intelligence will ever be 2 Nov 2026
- adding a new character: a taekwondo master that fights with fire I also updated the art style from 3d models to use generated sprites 1 Nov 2026
- the solution to economic doomerism is to be even more AGI-pilled I believe in a well-aligned ASI that can explain its work and bring us along, one that makes each individual maximally effective 30 Oct 2026
Link verified 23 Sept 2026. Recent items update automatically from the channel.
About the English under a post
Some people here publish in a language other than English. Where they do, this site shows a machine translation beneath the post, in this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the serif above is exactly what the person published, and it is what to quote them on. A translation can be wrong in ways that matter, especially about tone.
Only the post's own words are translated. A quoted post, a linked article and a belief on korrents.com are left in their original language.

