I Built the Same App With Claude Code and Codex
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Related videos
Claude Code's favorite tech stack
Theo - t3․gg
82.8k views
the end of Claude Code
Wes Roth
25.3k views
Claude Code got leaked
The PrimeTime
53.2k views
Claude Code Tutorial - Build Apps 10x Faster with AI
Programming with Mosh
95.2k views
Cursor, Claude Code and Codex all have a BIG problem
Theo - t3․gg
134.9k views
Claude Code - Full Tutorial for Beginners
Tech With Tim
56.5k views
The Only AI Coding Tools Worth Learning in 2026
Tech With Tim
38.1k views
Learning to code has changed
Tech With Tim
174.5k views
Lovable vs MeDo, which AI app builder actually works?
Tech With Tim
87.9k views
I Let 3 AIs Compete to Build the Same App…
Tech With Tim
4.3m views
Top Comments (10)
This lines up with my experience as well. Codex is just better at understanding large apps and maintaining them without breaking things, and taking instructions to do so. Claude is quicker at some tasks, but shittier to work with long term.
Antigravity was not harmed in this video
what if they copied each other because they are in the same folder
I would be interested to see your results doing the same thing in Cursor. I did a less scientific thing with a small project between Cursor and Claude. Cursor's eval of Claude was "wow, this is not great". Claude's analysis of Cursor was "hey, this is pretty good." They both had the context from their own creations and Claude basically said "Cursor did a better job". I have been a developer for over 30 years but I did not drill down and look at the actual code the way you did.
Great video, really liked the you tried the evaluate on different metrics instead of just 'vibing' who did better
Haha you're the first person I've seen use the same "ASCII" concept mock ups lolol, I do the EXACT same thing and it has always gotten me so much further than when I hadn't, it's silly effective! Cheers m808!
Click this link https://boot.dev/?promo=TECHWITHTIM and use my code TECHWITHTIM to get 25% off your first payment for boot.dev
Thanks for the video!
Great work Thank you The project is in GitHub?
For pretty complex SaaS projects with lots of integrations and moving parts in my office, Claude gives me better, straight-to-the-point fixes and features. It’s also more aware of how a new feature or bug fix might affect other parts of the system. I’m using the CLI version, and with Codex, changes often deviate from the original scope, introduce bugs in other areas, and create a lot of QA issues. I also end up needing many follow-up prompts just to fix the new issues caused by previous changes.
Unlock the Data Inside
Turn Videos into Knowledge
- Get FREE 10/day: transcripts, summaries, chats
- Chat with videos, export text & PDF
- $1 free API credit for RAG, chatbots & research
Free forever plan • All features unlocked
Top Comments (10)
This lines up with my experience as well. Codex is just better at understanding large apps and maintaining them without breaking things, and taking instructions to do so. Claude is quicker at some tasks, but shittier to work with long term.
Antigravity was not harmed in this video
what if they copied each other because they are in the same folder
I would be interested to see your results doing the same thing in Cursor. I did a less scientific thing with a small project between Cursor and Claude. Cursor's eval of Claude was "wow, this is not great". Claude's analysis of Cursor was "hey, this is pretty good." They both had the context from their own creations and Claude basically said "Cursor did a better job". I have been a developer for over 30 years but I did not drill down and look at the actual code the way you did.
Great video, really liked the you tried the evaluate on different metrics instead of just 'vibing' who did better
Haha you're the first person I've seen use the same "ASCII" concept mock ups lolol, I do the EXACT same thing and it has always gotten me so much further than when I hadn't, it's silly effective! Cheers m808!
Click this link https://boot.dev/?promo=TECHWITHTIM and use my code TECHWITHTIM to get 25% off your first payment for boot.dev
Thanks for the video!
Great work Thank you The project is in GitHub?
For pretty complex SaaS projects with lots of integrations and moving parts in my office, Claude gives me better, straight-to-the-point fixes and features. It’s also more aware of how a new feature or bug fix might affect other parts of the system. I’m using the CLI version, and with Codex, changes often deviate from the original scope, introduce bugs in other areas, and create a lot of QA issues. I also end up needing many follow-up prompts just to fix the new issues caused by previous changes.