Jason Sich

AI-native development. Outcome-first thinking. The frontier isn't where the danger is.

[Theo - t3.gg @theo]

Too many words.

$20 = 1x
$100 = 5x
$200 = 10x

$100 and $200 tiers are doubled til May 31st

[OpenAI @OpenAI]

We’re updating our ChatGPT Pro and Plus subscriptions to better support the growing use of Codex.

We’re introducing a new $100/month Pro tier. This new tier offers 5x more Codex usage than Plus and is best for longer, high-effort Codex sessions.

In ChatGPT, this new Pro tier still offers access to all Pro features, including the exclusive Pro model and unlimited access to Instant and Thinking models.

To celebrate the launch, we’re increasing Codex usage for a limited time through May 31st so that Pro $100 subscribers get up to 10x usage of ChatGPT Plus on Codex to build your most ambitious ideas.

[Jason Sich @SichDotDev]

@ryancarson

[Jason Sich @SichDotDev]

If you're struggling with the OpenAI models in OpenClaw try:

- OpenClaw 2026.4.5+
- GPT 5.4, thinking = high + fast = true
- Use an agent to clean up cruft from your agent files and align them with the 5.4 prompting guide.

[Sam Altman @sama]

To celebrate 3 million weekly codex users, we are resetting usage limits.

We will do this every million users up to 10 million.

Happy building!

[Jason Sich @SichDotDev]

If you’re new to OpenClaw check this out, I’ve made all of the same mistakes.

[Philipp Keller @philkellr]

I approached OpenClaw wrong

3 wrong assumptions I made:

#1 Openclaw knows how to build an openclaw agent
I assumed that the main agent would know (or inject) OpenClaw docs and best practices. It doesn't.

My agent guided me to add my own mobile number to WhatsApp. The docs prominently warn to not do this.

My agent motivated me to install a 3 level memory system with 3rd party tools: OpenClaw has their out of the box solution.

Of course I also never read the docs myself. So I left the architecture to my own agent, having no clue all the time what I'm building and what the best practice would be and when things didn't work out I had no plan on how to fix it. And no - I didn't take a cheap model. I always built with Sonnet or Opus.

I assumed OpenClaw was some magic device which figures everything out on its own. After months of vibecoding where I treated AI as junior dev I should have known better.

#2 My agent would keep my codebase clean
Throughout building my assistant I changed my mind how things should be built: I learned new concepts, gave up features mid way. I sprinkled in some "please clean up" prompts but the code base degraded fast and I slopped myself into a corner.

My markdown files became huge. I got code duplication. With every project I had before I'd do cleanup sessions, for whatever reason I didn't do this with OpenClaw.

#3 OpenClaw updates would be seamless
Also here I assumed things would just magically work: openclaw update and then running openclaw doctor and my agent would be back.

I never cared to read through the release notes.

I somehow didn't realize that the project was only 2 months old and that of course it would have breaking changes with every version.

---

I'm now reading through the docs. Frankly they are a bit a mess but I finally understand how openclaw is working. I'm getting my initial excitement back. I feel like the architect again and can help my agent to build itself using best practices.

[Marc Andreessen 🇺🇸 @pmarca]

Magical OpenClaw experiences that use frontier models cost $300-1,000/day today, heading to $10,000/day and more. The future shape of the entire technology industry will be how to drive that to $20/month.

[Jason Sich @SichDotDev]

@danshipper @every Well earned! Your contributions to industry have been 🤌

[Jason Sich @SichDotDev]

This is not the Hermes that you're looking for...

[Jason Sich @SichDotDev]

Continually impressed by Codex finding corner case bugs

> Runs a review with Codex
> Finds 1 issue
> "That can't be an issue", verifies things work fine in my browser
> Codex thinks... try again, but navigate directly to the page
> 😳 bug revealed

[Jason Sich @SichDotDev]

@Shpigford Working with AI is such a wild ride of emotions. One minute you're you can't believe how amazing it is and the very next you literally want to punch your computer.

[Jason Sich @SichDotDev]

Hopefully my wildest “dreams” now come true.

[OpenClaw🦞 @openclaw]

OpenClaw 2026.4.5 🦞

🎬 Built-in video + music generation
🧠 /dreaming is now real
🔀 Structured task progress
⚡ Better prompt-cache reuse
🌍 Control UI + Docs now speak 12 more languages

Anthropic cut us off. GPT-5.4 got better. We moved on. https://t.co/T3LaSJYOvU

[Jason Sich @SichDotDev]

@Shpigford High + Fast is the way to go.

[Theo - t3.gg @theo]

Claude Code now throws an error if you use it to try and analyze the Claude Code source

[Jason Sich @SichDotDev]

OpenAI models in OpenClaw have the personality of a Fortune 500 middle manager by default.

Excited to see how this and the other upcoming changes fix that.

[OpenClaw🦞 @openclaw]

Give your agent the Molty spicy SOUL upgrade: stronger opinions, less corporate sludge, better instincts, and replies that might finally sound like someone worth talking to at 2am 🦞https://t.co/Is4Q1GsMpQ

[Jason Sich @SichDotDev]

@danshipper Easy to miss, but makes a huge difference. Ant defaults to “adaptive” thinking, really wish OpenAI had that option.

[Jason Sich @SichDotDev]

@Shpigford Where have you been all my life?!

[Peter Steinberger 🦞 @steipete]

you said GPT felt like bread, so we stopped serving carbs. 🥖❌🦞

[Peter Steinberger 🦞 @steipete]

@keithtyser @openclaw Found that issue and fixing it, next release will work better with GPT.

[Jason Sich @SichDotDev]

If your agent doesn't talk to you like this YNGMI