Offshore
Photo
Michael Fritzell (Asian Century Stocks)
RT @CompoundingUp: Ppl really be paying tuition to attend an esports college
2415 Human Holdings https://t.co/ubm7gEk9yR
tweet
RT @CompoundingUp: Ppl really be paying tuition to attend an esports college
2415 Human Holdings https://t.co/ubm7gEk9yR
tweet
Michael Fritzell (Asian Century Stocks)
RT @finphysnerd: @capytalmgmt I think a good way to avoid stuff like this in EM is to focus on companies that are returning capital. Looks like Huisen only had 1 dividend in 2020. It's basically the first thing I look at in cheap Chinese small caps
tweet
RT @finphysnerd: @capytalmgmt I think a good way to avoid stuff like this in EM is to focus on companies that are returning capital. Looks like Huisen only had 1 dividend in 2020. It's basically the first thing I look at in cheap Chinese small caps
tweet
Javier Blas
Trump on Iran:
“I think a deal can't be reached […] We have plenty of time. If you remember Venezuela, we waited around for a while, and we're in no rush. We have very good talks going with Iran [….] We're going to meet again early next week, and they want to make a deal.”
tweet
Trump on Iran:
“I think a deal can't be reached […] We have plenty of time. If you remember Venezuela, we waited around for a while, and we're in no rush. We have very good talks going with Iran [….] We're going to meet again early next week, and they want to make a deal.”
tweet
Offshore
Photo
God of Prompt
RT @godofprompt: Stop telling LLMs like Claude and ChatGPT what to do.
Start asking them questions instead.
I replaced all my instruction prompts with question prompts.
Output quality: 6.2/10 → 9.1/10
This is called "Socratic prompting" and here's how it works: https://t.co/UWgQXYHS0D
tweet
RT @godofprompt: Stop telling LLMs like Claude and ChatGPT what to do.
Start asking them questions instead.
I replaced all my instruction prompts with question prompts.
Output quality: 6.2/10 → 9.1/10
This is called "Socratic prompting" and here's how it works: https://t.co/UWgQXYHS0D
tweet
God of Prompt
🚨 Wait. Did Greg just soft-announce the end of traditional coding?
“By March 31st, for any technical task, the tool of FIRST RESORT is interacting with an agent rather than using an editor or terminal.”
That’s OpenAI’s internal mandate.
Their own engineers stopped writing code in December. Codex writes it. They debug, review, guide.
The 6-step framework isn’t “how to try AI tools.” It’s “how we’re restructuring engineering teams around agents as the default.”
Key signals:
→ AGENTS[.md] becomes standard project documentation
→ Skills directories (reusable agent workflows)
→ “Agents captain” role on every team
→ Codebases restructured to be “agent-first”
This is OpenAI showing us their internal playbook for a transition they believe is inevitable.
If the company building the models doesn’t use editors anymore… what does that tell you about 2026?
Software development is undergoing a renaissance in front of our eyes.
If you haven't used the tools recently, you likely are underestimating what you're missing. Since December, there's been a step function improvement in what tools like Codex can do. Some great engineers at OpenAI yesterday told me that their job has fundamentally changed since December. Prior to then, they could use Codex for unit tests; now it writes essentially all the code and does a great deal of their operations and debugging. Not everyone has yet made that leap, but it's usually because of factors besides the capability of the model.
Every company faces the same opportunity now, and navigating it well — just like with cloud computing or the Internet — requires careful thought. This post shares how OpenAI is currently approaching retooling our teams towards agentic software development. We're still learning and iterating, but here's how we're thinking about it right now:
As a first step, by March 31st, we're aiming that:
(1) For any technical task, the tool of first resort for humans is interacting with an agent rather than using an editor or terminal.
(2) The default way humans utilize agents is explicitly evaluated as safe, but also productive enough that most workflows do not need additional permissions.
In order to get there, here's what we recommended to the team a few weeks ago:
1. Take the time to try out the tools. The tools do sell themselves — many people have had amazing experiences with 5.2 in Codex, after having churned from codex web a few months ago. But many people are also so busy they haven't had a chance to try Codex yet or got stuck thinking "is there any way it could do X" rather than just trying.
- Designate an "agents captain" for your team — the primary person responsible for thinking about how agents can be brought into the teams' workflow.
- Share experiences or questions in a few designated internal channels
- Take a day for a company-wide Codex hackathon
2. Create skills and AGENTS[.md].
- Create and maintain an AGENTS[.md] for any project you work on; update the AGENTS[.md] whenever the agent does something wrong or struggles with a task.
- Write skills for anything that you get Codex to do, and commit it to the skills directory in a shared repository
3. Inventory and make accessible any internal tools.
- Maintain a list of tools that your team relies on, and make sure someone takes point on making it agent-accessible (such as via a CLI or MCP server).
4. Structure codebases to be agent-first. With the models changing so fast, this is still somewhat untrodden ground, and will require some exploration.
- Write tests which are quick to run, and create high-quality interfaces between components.
5. Say no to slop. Managing AI generated code at scale is an emerging problem, and will require new processes and conventions to keep code quality high
- Ensure that some human is accountable for any code that gets merged. As a code reviewer, maintain at least the same bar as you would for human-written code, and make sure the author u[...]
🚨 Wait. Did Greg just soft-announce the end of traditional coding?
“By March 31st, for any technical task, the tool of FIRST RESORT is interacting with an agent rather than using an editor or terminal.”
That’s OpenAI’s internal mandate.
Their own engineers stopped writing code in December. Codex writes it. They debug, review, guide.
The 6-step framework isn’t “how to try AI tools.” It’s “how we’re restructuring engineering teams around agents as the default.”
Key signals:
→ AGENTS[.md] becomes standard project documentation
→ Skills directories (reusable agent workflows)
→ “Agents captain” role on every team
→ Codebases restructured to be “agent-first”
This is OpenAI showing us their internal playbook for a transition they believe is inevitable.
If the company building the models doesn’t use editors anymore… what does that tell you about 2026?
Software development is undergoing a renaissance in front of our eyes.
If you haven't used the tools recently, you likely are underestimating what you're missing. Since December, there's been a step function improvement in what tools like Codex can do. Some great engineers at OpenAI yesterday told me that their job has fundamentally changed since December. Prior to then, they could use Codex for unit tests; now it writes essentially all the code and does a great deal of their operations and debugging. Not everyone has yet made that leap, but it's usually because of factors besides the capability of the model.
Every company faces the same opportunity now, and navigating it well — just like with cloud computing or the Internet — requires careful thought. This post shares how OpenAI is currently approaching retooling our teams towards agentic software development. We're still learning and iterating, but here's how we're thinking about it right now:
As a first step, by March 31st, we're aiming that:
(1) For any technical task, the tool of first resort for humans is interacting with an agent rather than using an editor or terminal.
(2) The default way humans utilize agents is explicitly evaluated as safe, but also productive enough that most workflows do not need additional permissions.
In order to get there, here's what we recommended to the team a few weeks ago:
1. Take the time to try out the tools. The tools do sell themselves — many people have had amazing experiences with 5.2 in Codex, after having churned from codex web a few months ago. But many people are also so busy they haven't had a chance to try Codex yet or got stuck thinking "is there any way it could do X" rather than just trying.
- Designate an "agents captain" for your team — the primary person responsible for thinking about how agents can be brought into the teams' workflow.
- Share experiences or questions in a few designated internal channels
- Take a day for a company-wide Codex hackathon
2. Create skills and AGENTS[.md].
- Create and maintain an AGENTS[.md] for any project you work on; update the AGENTS[.md] whenever the agent does something wrong or struggles with a task.
- Write skills for anything that you get Codex to do, and commit it to the skills directory in a shared repository
3. Inventory and make accessible any internal tools.
- Maintain a list of tools that your team relies on, and make sure someone takes point on making it agent-accessible (such as via a CLI or MCP server).
4. Structure codebases to be agent-first. With the models changing so fast, this is still somewhat untrodden ground, and will require some exploration.
- Write tests which are quick to run, and create high-quality interfaces between components.
5. Say no to slop. Managing AI generated code at scale is an emerging problem, and will require new processes and conventions to keep code quality high
- Ensure that some human is accountable for any code that gets merged. As a code reviewer, maintain at least the same bar as you would for human-written code, and make sure the author u[...]
Offshore
God of Prompt 🚨 Wait. Did Greg just soft-announce the end of traditional coding? “By March 31st, for any technical task, the tool of FIRST RESORT is interacting with an agent rather than using an editor or terminal.” That’s OpenAI’s internal mandate. Their…
nderstands what they're submitting.
6. Work on basic infra. There's a lot of room for everyone to build basic infrastructure, which can be guided by internal user feedback. The core tools are getting a lot better and more usable, but there's a lot of infrastructure that currently go around the tools, such as observability, tracking not just the committed code but the agent trajectories that led to them, and central management of the tools that agents are able to use.
Overall, adopting tools like Codex is not just a technical but also a deep cultural change, with a lot of downstream implications to figure out. We encourage every manager to drive this with their team, and to think through other action items — for example, per item 5 above, what else can prevent a lot of "functionally-correct but poorly-maintainable code" from creeping into codebases. - Greg Brockman tweet
6. Work on basic infra. There's a lot of room for everyone to build basic infrastructure, which can be guided by internal user feedback. The core tools are getting a lot better and more usable, but there's a lot of infrastructure that currently go around the tools, such as observability, tracking not just the committed code but the agent trajectories that led to them, and central management of the tools that agents are able to use.
Overall, adopting tools like Codex is not just a technical but also a deep cultural change, with a lot of downstream implications to figure out. We encourage every manager to drive this with their team, and to think through other action items — for example, per item 5 above, what else can prevent a lot of "functionally-correct but poorly-maintainable code" from creeping into codebases. - Greg Brockman tweet
Offshore
Video
God of Prompt
RT @alex_prompter: OpenClaw broke the internet
But you DON'T need to setup any servers to use it
Here's the easiest way to run OpenClaw on a website
No Mac Minis required https://t.co/6xspOtHaxT
tweet
RT @alex_prompter: OpenClaw broke the internet
But you DON'T need to setup any servers to use it
Here's the easiest way to run OpenClaw on a website
No Mac Minis required https://t.co/6xspOtHaxT
tweet
Offshore
Photo
God of Prompt
RT @godofprompt: Claude Opus 4.6 is a monster.
I just used it for:
- automating marketing tasks
- building full websites and apps
- writing viral X threads, LinkedIn posts, and YouTube scripts
And it did all this in minutes.
Here are 10 prompts you can steal to unlock its full potential: https://t.co/bpXZy0CfyK
tweet
RT @godofprompt: Claude Opus 4.6 is a monster.
I just used it for:
- automating marketing tasks
- building full websites and apps
- writing viral X threads, LinkedIn posts, and YouTube scripts
And it did all this in minutes.
Here are 10 prompts you can steal to unlock its full potential: https://t.co/bpXZy0CfyK
tweet
Offshore
Video
God of Prompt
RT @alex_prompter: this is your competition https://t.co/m7XgDebbQW
tweet
RT @alex_prompter: this is your competition https://t.co/m7XgDebbQW
OpenClaw broke the internet
But you DON'T need to setup any servers to use it
Here's the easiest way to run OpenClaw on a website
No Mac Minis required https://t.co/6xspOtHaxT tweet