Edgar García

4.1K posts

Edgar García banner
Edgar García

Edgar García

@edgaragr

CTO & Co-founder Building AI products for real estate investors and developers | Tech × Product × Business ✝ https://t.co/oeuwMFvfqr | https://t.co/nyEhfcke3K

Dominican Republic Katılım Aralık 2009
572 Takip Edilen204 Takipçiler
Edgar García retweetledi
ClaudeDevs
ClaudeDevs@ClaudeDevs·
Useful tip to cut time-to-first-token on longer prompts in the API: pre-warm the prompt cache. Send your system prompt before the user prompt. Claude writes it to the cache, but skips generating any output. When the real user request lands, it'll hit a warm cache.
ClaudeDevs tweet media
English
94
184
3K
269.8K
Edgar García
Edgar García@edgaragr·
Codex, please, I would like to continue monitoring and interacting with my local codex sessions as we do with Claude 🙏
English
0
0
0
13
Edgar García
Edgar García@edgaragr·
@trq212 @bcherny What about token usage? I think one of the benefits of using Markdown is its simplicity. For example, if we get a plan as HTML and later need to use it as a reference for another task, wouldn’t that be more costly?
English
0
0
1
643
Edgar García
Edgar García@edgaragr·
Now it's turn to OpenAi, show us what you have :) Codex for mobile ?
English
0
0
0
23
Edgar García retweetledi
Boris Cherny
Boris Cherny@bcherny·
Two big changes for Claude Code today: 5-hour rate limits doubled on Pro, Max, Team, and seat-based Enterprise, and we’ve removed the peak-hours limits for Pro and Max. Both effective today.
English
102
38
1.5K
73.8K
Edgar García
Edgar García@edgaragr·
IMO this is the biggest new and surprise, didn’t see it coming. Now let’s see how fast they can start using it. We need it ASAP.
Claude@claudeai

Our agreement with @SpaceX means we will use all the compute capacity at their Colossus 1 data center. This will give us over 300 megawatts of additional capacity to deploy within the month.

English
0
0
0
15
Edgar García retweetledi
Claude
Claude@claudeai·
We’ve agreed to a partnership with @SpaceX that will substantially increase our compute capacity. This, along with our other recent compute deals, means that we’ve been able to increase our usage limits for Claude Code and the Claude API.
English
4.8K
12.1K
131K
23.7M
Edgar García
Edgar García@edgaragr·
@nickchapsas I did that and sadly we don’t have auto-mode nor xhigh thinking 🫠 that’s the tradeoff.
English
0
0
1
1.9K
Nick Chapsas
Nick Chapsas@nickchapsas·
claude - - model claude-opus-4-6, thank me later
English
33
6
211
37.4K
Edgar García
Edgar García@edgaragr·
@Jaytel @MarceloRet41877 It is so slow that most of the time I think I can do it faster by myself. This is not that we paid for initially.
English
0
0
0
60
Jaytel
Jaytel@Jaytel·
4.7 is completely unusable
English
449
173
5.9K
1M
IndiJo
IndiJo@odd_joel·
@edgaragr @sama Moshi covers the codex-remote-control gap on iphone — ssh into your machine and the Mosh protocol keeps the session alive through wifi switches and sleep, plus push notifs when codex stalls for approval so you can unblock from anywhere
English
1
0
0
129
Edgar García
Edgar García@edgaragr·
The only thing I miss from Claude Code is the mobile app, @sama please make it real soon 🙏 codex on desktop is so nice but it miss a way to control it remotely.
English
1
0
0
114
Edgar García
Edgar García@edgaragr·
@shub0414 Yep, unpopular opinion but 100% right, probably by EOY we will have a price bump or early next year. Right now OpenAI and Anthropic are attracting people but we know the price we are paying isn't proportional to the cost of the investment these companies are doing.
English
0
0
1
49
Shub
Shub@shub0414·
Unpopular opinion: Your $20/month AI subscription is a lie. You’re living on VC handouts. OpenAI and Anthropic are burning billions to subsidize your productivity. That money will run out. When the subsidies dry up, the cost of "intelligence" will skyrocket. The play? Build your agents and workflows now while compute is pennies on the dollar. Lock in your advantage before the price tag catches up to the tech.
English
121
13
238
21.9K
Edgar García
Edgar García@edgaragr·
@thsottiaux On status page I would like to know what is consuming my tokens, for example, which skills are consuming more tokens, etc.
English
0
0
0
100
Tibo
Tibo@thsottiaux·
What are we obviously not getting right with Codex?
English
2.8K
29
2.5K
611K
Vishal
Vishal@vishaltweetup·
If AI writes 90% of the code, what do we call developers now?
English
124
4
86
7.1K
Edgar García
Edgar García@edgaragr·
@thsottiaux A rewind equivalent command to remove the last thing from the context window.
English
0
0
4
296
Edgar García
Edgar García@edgaragr·
Hey Vibe coders which option do you think is the best one right now and why? I'm talking about plans: 1. Claude $100 - Codex $100 2. Claude $200 - Codex $20 3. Codex $200 4. Claude $200
English
0
0
0
29

Keşfet

@AnthropicAI @OpenAI @yasser_elsaid_ @trq212 @bcherny @SpaceX @nickchapsas @Jaytel