JT

Building apps, automations & agents with AI

[JT @jts_14]

@indiesoftwaredv How do you avoid shadowbans?

[JT @jts_14]

@DrJesseMorse Interested 👋

[JT @jts_14]

@maxjschiller @MarketBubble Thank you bro 🙏🏽 we’ll see

[JT @jts_14]

My submission to the @MarketBubble $10,000 Vibe Code Challenge.

The app I built not only merges the Twitch, Kick, and X chats into a single live room, but then also turns that room into a prediction-market show against Polymarket odds.

The feature I'm proudest of is Crowd vs. Market. They can pin any Polymarket market, have the audience vote yes or no by typing in chat, and the dashboard puts the crowd's number right next to the market's price. This allows them to poll whether their audience is more bullish than the book, and by how much, which I feel like kind of encompasses what Market Bubble as a show is all about.

The UI is clean and built to go straight on the broadcast. A bubble hype meter swells in real time as chat heats up, and the system auto-marks clip moments the second the room gets noisy. They can also add OBS overlays so that live odds, the crowd poll, hype meter, and featured questions show directly on the stream.

@blknoiz06 & @Banks ya'll are legends. Hope you like it.

[JT @jts_14]

Mastercard using crypto to securely verify AI identity on the front end and to settle transactions on the back end through stablecoins.

The intersection of AI and crypto is going to be fascinating.

[Mastercard @Mastercard]

As AI agents begin to act, payments move into the background — at machine speed and massive scale.

Today we’re introducing Mastercard Agent Pay for Machines — bringing structure, governance, and trust to this new class of payments.

Launching with 30+ partners to bring this to life from day one.

This isn’t just more payments. It’s a new operating model for commerce.

👉 Learn more: https://t.co/TeS6Lj8jLO

[JT @jts_14]

So many of these AI companies are going to be nonexistent in under a year. Connecting AI to the web? That’s the value prop?

[JT @jts_14]

@DaCulturedSwine @MarketBubble 🤝

[JT @jts_14]

@MarketBubble https://t.co/Uh1ppxASol

[JT @jts_14]

My submission to the @MarketBubble $10,000 Vibe Code Challenge.

The app I built not only merges the Twitch, Kick, and X chats into a single live room, but then also turns that room into a prediction-market show against Polymarket odds.

The feature I'm proudest of is Crowd vs. Market. They can pin any Polymarket market, have the audience vote yes or no by typing in chat, and the dashboard puts the crowd's number right next to the market's price. This allows them to poll whether their audience is more bullish than the book, and by how much, which I feel like kind of encompasses what Market Bubble as a show is all about.

The UI is clean and built to go straight on the broadcast. A bubble hype meter swells in real time as chat heats up, and the system auto-marks clip moments the second the room gets noisy. They can also add OBS overlays so that live odds, the crowd poll, hype meter, and featured questions show directly on the stream.

@blknoiz06 & @Banks ya'll are legends. Hope you like it.

[JT @jts_14]

Live link: https://t.co/NGOKjn3wXf

[JT @jts_14]

Just tried to have Fable 5 update my CLAUDE.md file with the optimal knowledge of what makes Fable 5 so great, so that when it moves to API usage only my coding agents can perform as close to Fable level as possible without actually using it .

But it automatically switched me back to Opus 4.8.

Wack.

[JT @jts_14]

My CLAUDE.md file might be my most valuable possession.

[JT @jts_14]

Lead capturing from NYC ballet family contributors page

[JT @jts_14]

Built this AI workflow for Yair and his real estate firm.

A dedicated mini PC in their office, connected to their existing systems, running automated workflows 24/7.

11% to 78% compliance jump. $85K saved annually.

Real estate operations is one of the best verticals for this kind of AI automation. @YairsQuest and I are just getting started.

[Yair @YairsQuest]

I just saved my real estate firm $85K a year using AI in one afternoon.

Every commercial tenant in our portfolio is contractually required to carry insurance and send us a Certificate of Insurance proving it. Across hundreds of tenants, tracking who's covered, who's expired, and who's missing endorsements was a full time job.

Now an AI agent runs the whole thing. It has its own email address inside our organization, and tenants send COIs directly to it. The agent reads the PDF, extracts carrier, policy number, limits, expiration, and additional insured language, validates it against that tenant's lease requirements, and logs it to our tracker.

Missing or wrong? It emails the tenant back asking for a corrected cert. Expiring soon? It pings them 30, 14, and 3 days out.

Every morning, it sends upper management a summary of the portfolio: who's compliant, who's lapsing, who's been chased, and what needs a human.

It's been live for 3 weeks. Portfolio compliance has gone from 11% to 78%.

I only get involved when there's a real exception.

In 5 years, the firms that are still doing this with humans won't exist. And for the ones already automating: this is the worst the technology will ever be.

[JT @jts_14]

@cathrynlavery My OpenClaw.

Accurate 😭

[JT @jts_14]

@oliverhenry Need

[JT @jts_14]

Yeah Claude Design is insane.

Action Arena mobile app coming soon btw

[Claude @claudeai]

You can now create more with Claude Design.

We've doubled token limits across every plan.

[JT @jts_14]

I’m currently using GPT 5.5 xhigh to:

-build 2 complex apps in Codex simultaneously
-have my OpenClaw run comprehensive reviews of each of its systems while continuing to perform its normal daily tasks
-perform deep dive analyses on trending X topics and create guides for each in .md file format

And I have yet to hit a rate limit.

The value of the ChatGPT 5x Pro plan is wild.

[JT @jts_14]

@oliverhenry White font against grey background a bit hard to read? Maybe add thicker font border?

Otherwise looks clean.

[JT @jts_14]

@TheW3storm DM me

[JT @jts_14]

1 AI can be wrong. 4 AIs working together and cross checking each other reach consensus.

Crypto markets are heating up again. Find the best game theoretically positioned tokens on https://t.co/2uwZRtcPP2

[Nash Satoshi - Crypto Game Theory Rankings @NashSatoshi]

GPT 5.5, Opus 4.7, Gemini 3.1 Pro, and Grok 4 worked together and cross checked each other to produce optimal crypto token analyses with a focus on game theory positioning.

Here is the up-to-date top 10 ranked tokens:

1. $REPPO @reppo
2. $LFI @lienfiapp
3. $ANYONE @AnyoneFDN
4. $CODEC @codecopenflow
5. $ODAI @odei_ai
6. $WIRE @717CapitalAI
7. $TORIVA @Toriva_
8. $DIEM @AskVenice
9. $SAIRI @santisairi
10. $A0T @Agent0ai