Izzy

In the heart of every rebel burns a flame that refuses to be tamed. Defiant, wild, and forever alive.

[Izzy @iiizzy]

https://t.co/QEl0OymHCj

[Izzy @iiizzy]

My personal site costs $0 a month to run.
No database. No plugin updater. No login page getting brute-forced.
Static pages on Cloudflare. Git is the CMS. One tiny API for views, reactions and comments.
Lighthouse: 99–100 across the board.

[Izzy @iiizzy]

@mathiasvrb Me too 😁 I was betting on the opposite before trying

[Izzy @iiizzy]

I cut my MCP server's tool list from 88 to 4 and measured 334 Claude Code sessions.
Context per session dropped 12x.
Cost went UP 65%.
Prompt caching had already made the big tool list cheap. Routing swapped cheap tokens for expensive ones.
Link below 👇

[Izzy @iiizzy]

https://t.co/0OzFi2tpur

[Izzy @iiizzy]

Jev isn't an LLM. TypeSafe calls it a System One model: send it state and typed questions, get typed decisions back with a probability each. No text generation.
I send it 134 questions per MCP request. 1 HTTPS call. 651 ms median. $0.0005.
It's my tool router now. 👇

[Izzy @iiizzy]

https://t.co/0OzFi2tpur

[Izzy @iiizzy]

Meet Stripe's Knowledge AI Platform https://t.co/2W2AsXoXeu

[Izzy @iiizzy]

Your next buyer might never load your site. Their agent will.

Vercel traced 1,033 agent runs. Only two things stopped agents from reading a page: JavaScript-only content and 403s to agent traffic.
https://t.co/IQpN7HLRr8

[Izzy @iiizzy]

@FishAudio Fish

[Fish Audio @FishAudio]

Today we’ve raised $52M Seed and we are announcing the public launch of S2.1 Pro.

>It can clone a voice from 5 seconds of audio
>2x faster than Cartesia & 1/6th the cost of Eleven Labs
>most expressive model with word level control over emotion, intonation, pacing etc

We support frontier AI companies including HeyGen, LiveKit, Retell, Sanas, and OpenArt all run our model in production.

If you're a business and we can't cut your voice AI costs by 50%, we'll give you 1 year of Fish Audio for free.
Book a demo: https://t.co/vHkyZf9JoG

To celebrate our first birthday, we'll give you 1 month of S2.1 Pro for free. Like, retweet, and comment “Fish” to get it.

[Izzy @iiizzy]

The best AI agent on Earth just scored 3.75% on real freelance work
Same models score 90%+ on MMLU
The Remote Labor Index tested agents on 240 real Upwork projects of actual paid work. Almost all of it failed
If your AI strategy cites benchmarks, read 👇
https://t.co/nEJqHEyP4z

[Izzy @iiizzy]

@haukejung @X Hello 👋 https://t.co/3e2RHEnQZc

[Izzy @iiizzy]

AI didn't make our work faster; it changed its shape. We launched a company in a week, no relay between research, strategy, copy, design, and build. It's not the tools. It's the shape.

https://t.co/cFcUr1bWfk

[Izzy @iiizzy]

🛠️ The good news? You don't need to master a dozen tools. You only need three to unlock 80% of a Claude Code-powered RevOps workflow:

- An agent memory layer

- A stealth browser

- A solid skills library

[Izzy @iiizzy]

🧠 The biggest bottleneck in RevOps? Human memory. 82% of revenue leaders admit half their institutional knowledge is stuck in people's heads.

Worse? Out-of-the-box Claude Code forgets everything when you close the tab. Enter the rise of "agent memory" repos.

[Izzy @iiizzy]

🔮 Want to see the future of RevOps? Look at trading and DeFi repos today.

They are perfecting autonomous "observe-decide-act" loops in the open.

Give it two years, and this exact agent-native architecture will be the standard engine running revenue operations.