vibe-coding | co-founder github.com/Taraxa-project | director @ Monitor Deloitte | founder @ IoT & health startups in China | EE @Stanford
Joined March 2009
- Tweets 3,209
- Following 255
- Followers 3,165
- Likes 3,903
233 Photos and videos
Steven Pu retweeted
Jun 12
The next war won't be won by armies, navies or air forces alone.
It'll be won by the country whose 19 year olds can code, whose factories can build drones in weeks not years, and whose grid stays on when someone tries to switch it off.
Industry. Society. Economy. That's the fight now.
We're not ready. And we're not being honest about what getting ready will cost.
1,118
2,306
13,403
760,046
Absolutely nuts. I had literally thought about why something like this doesn't exist yet ~5 days ago. If this does just a fraction of what it claims, it'd blow platforms like Roblox out of the water.
Jun 11
Meet Higgsfield Games.
For the first time, build and deploy multiplayer games from one prompt, in any genre, 2D or 3D, with best-in-class characters, props, and settings generated by Higgsfield MCP.
Powered by Claude Fable 5.
Try on Claude via MCP and on our Supercomputer.
152
Wow, very first time I see @GeminiApp cite @grok grokpedia. Cited multiple times in the same session too, with no Wikipedia citations at all, when it contained the same information entries.
Maybe grokpedia performs better with agents?
1
56
LLMs perfectly understands misspelled words, poorly constructed phrases, mis-labeled concepts, a whole host of things that it just auto-corrects.
Hate to dunk on a builder but this is really not that useful.
Jun 10
I kept spotting typos in my Claude Code prompts after I'd sent them. So I built tuipo β Grammarly for your terminal
It underlines typos as you type in any TUI β CC, Codex, Aider, vim, your shell β and never touches the app it wraps.
github.com/ARahim3/tuipo
@steipete @bcherny
80
Steven Pu retweeted
Jun 10
only boomers fix typos in prompts. llms perfectly understand you even if you mistype.
358
195
5,296
160,154
tackit 0.7 -> 0.8.1, more dog-fooding, highlights:
filtering for ls/board ops for efficiency
bulk link creation & resolution
FTS query sanitation escapes special chars
skill against semantic links & book-keeping
-- trimmed summaries from ls/board to reduce bloat
1
1
1
182
Could just be people building their own personal apps, and donβt want to deal with janky TestFlight.
163
Looking at my timeline, this seems appropriate.
music.youtube.com/watch?v=48β¦
104
AI is fighting back HARD against behavioral modification.
It largely stopped creating tasks that are indices of other tasks, but, it snuck this behavior back by burying semantic (instead of graph edges like it's supposed to) dependencies inside task descriptions.
Forced me to write another rule in skills. I respect the AI's game. π«‘
82
Skill issues at big company means small new ones can eat their lunch
Jun 5
If big companies can't make a net return on their LLM token costs, that doesn't mean it's impossible to. In fact this is exactly what you'd expect to happen with a new technology. Incumbents can't use it well, and are replaced by upstarts who can.
60
29
367
53,751