paradite_ Profile Banner
Zhu Liang Profile
Zhu Liang

@paradite_

Followers
691
Following
3K
Media
504
Statuses
2K

You're absolutely right! You're absolutely right! You're absolutely right!

Singapore
Joined August 2009
Don't wanna be here? Send us removal request.
@paradite_
Zhu Liang
2 months
Finished testing Claude Opus 4 and Claude Sonnet 4 on my personal eval set. I am VERY impressed. Claude Opus 4 absolutely dominated other models in both coding and writing tasks. It is the best performing model for all 4 tasks given. It is worth nothing this model is very
Tweet media one
15
23
272
@paradite_
Zhu Liang
4 hours
Are you okay, Claude? Anthropic API appears to be overloaded 529 for me across different apps. Anthropic status page needs to be updated.
Tweet media one
Tweet media two
Tweet media three
1
0
2
@paradite_
Zhu Liang
5 hours
Looks like the Claude Code "nerf" is here. Some personal observations:. - No more automatic TODOs for small tasks (so far).- Reading files 10/15 lines at a time instead of 50/100 lines.- Failing at simple tasks.
1
0
0
@paradite_
Zhu Liang
6 hours
What's the strategy for AI coding companies are? What's Cursor gonna do next? What's Devin going build after acquiring Windsurf?.
0
0
0
@paradite_
Zhu Liang
6 hours
I think Claude Code is "enough" for me. I don't really care if it is 100% correct or reliable. I am happy to handle 10% edge cases myself. Going from time saving of 5 hours a day to 5.5 hours is nothing compared to previous generational leaps.
1
0
1
@paradite_
Zhu Liang
6 hours
2024, 70%: Cursor became really good in 2024. Edit multiple files. Run autonomously for 5 minutes. 2025, 80%: Claude Code came in 2025. No embeddings or codebase indexing. Uses grep and cli to navigate the codebase.
1
0
0
@paradite_
Zhu Liang
6 hours
AI coding is hitting diminishing return. 2021, 30%: GitHub Copilot was first AI coding tool. Autocomplete felt like magic. 2023, 60%: ChatGPT and GPT-4 was next. Huge intelligence jump. AI can actually write code given a task.
1
0
1
@paradite_
Zhu Liang
19 hours
Wtf Windsurf is joining Devin (Cognition).
Tweet media one
2
0
2
@paradite_
Zhu Liang
1 day
And it does work! I asked Claude Code to write a test with assertion and it works!
Tweet media one
1
0
0
@paradite_
Zhu Liang
1 day
Watching Claude Code trying to bypass bot detection on xAI website is really fun. After several failed attempts messing with UA and various headers, Claude Code resorted to using archive. org instead.
Tweet media one
1
0
1
@paradite_
Zhu Liang
1 day
Interesting strategy by Amp to use unix timestamp as version number. This coincidentally solves the 9.9 vs 9.11 problem that AI models are unfortunately being RLed to memorize as maths problem instead of software version number. Using the unix timestamp ensures that the version.
@chrisbbh
Christian Bager Bach Houmann
2 days
I was wondering why, every time I open my IDE, there's an Amp update. Nice pace from the Amp team. Screenshot taken on a Sunday :). No wonder it's such a good agent
Tweet media one
0
0
2
@paradite_
Zhu Liang
2 days
Some new information on Windsurf employees.
@nmasc_
natasha mascarenhas
2 days
Updated our Windsurf/Google story with new deal details: . - Employees with vested shares will receive cash.– Employees who joined less than 12 months ago are not vested and won’t get payouts under current terms.– Windsurf negotiated to keep $100M+ on its balance sheet; company.
0
0
2
@paradite_
Zhu Liang
2 days
For those curious, the full replay vod of the livestream is available here:.
0
0
1
@paradite_
Zhu Liang
2 days
So you spend less time working, while getting more work done (assuming agents are as effective as you). The efficiency and productivity gain of AI coding comes from the async nature of coding and the reduced engagement time for human to be present.
1
0
1
@paradite_
Zhu Liang
2 days
The other way to look at it from a productivity perspective: . Instead of spending 5 hours a day engaged in coding, you can now spend 1 hour in the morning and 1 hour in the afternoon prompting and reviewing, to get 6 hours of agent working time.
1
0
0
@paradite_
Zhu Liang
2 days
I could go read a book, write a blog post, or watch some YouTube videos and then come back to review the code. The effective time I have to spend being engaged with the coding process is reduced to around 30 minutes from 90 minutes, which gives a 66% time saving for me.
1
0
1
@paradite_
Zhu Liang
2 days
However, if I were to code it myself, I would have spent 90 minutes staring at the screen and spending my brain cells on solving bugs. Compare this with using Claude Code where I didn't have to be engaged after giving the initial prompt and before reviewing the results.
1
0
2
@paradite_
Zhu Liang
2 days
Some viewers pointed out that, it would probably taken me (experienced dev) similar amount of time to write code manually compared to using Claude Code. That's true, in terms of wall time (how much time it takes for a clock to run).
1
0
0
@paradite_
Zhu Liang
2 days
Just finished my first AI coding livestream with Claude Code, where I shipped 3 small-medium features to 2 apps across 4 repos in 90 minutes live on camera, without writing a single line of code. Some interesting observations on wall time vs human engagement time:
Tweet media one
3
2
13
@paradite_
Zhu Liang
3 days
So it's entirely logical to assume that we are no different from LLMs on a conceptual level. And we do in fact live in a simulation.
0
0
0
@paradite_
Zhu Liang
3 days
Humans have both system 1 thinking (automatic) and system 2 thinking (rational), similar to how LLMs would take parallel paths to generate tokens, one memorized shortcut response and one from rational deduction.
1
0
0