CEO & Co-Founder at @CLRTstudio I build, ship and teach Agentic AI.
Dubai, United Arab Emirates
Joined October 2023
- Tweets3.7K
- Following2.1K
- Followers1.8K
- Likes18.2K
Pinned Tweet
Reviewers caught 82% of swapped digits. Change the meaning instead, and the result gets uncomfortable.
1
1
43
Uber's annual AI budget was gone in four months, with roughly 5,000 engineers on the tools. It wasn't overspending, exactly. The budget was pricing the wrong thing.
12
5
542
Is Google's strongest AI model really this cheap? 💸 Gemini 4 Carbon costs $2 in and $10 out per million tokens until 30 November, then it doubles to $4 and $20, per Google's pricing page. For founders and operators setting next year's AI budget.
2
3
429
It started with a Reddit comment behind a spoiler tag. It ended with an AI browser handing over a one-time passcode.
3
6
118
The full piece: clrtstudio.com/insights/brow…
10
GPT-4o's 128,000-token window held up to about 8,000 in one test. The brochure measures what fits. Your contract review needs a different test: can it find the clause you need when the question uses different words?
Would you give every task to the $40 model? 🤔 Fable 5.2 finished 41.3% of real freelance projects at a pro standard, yet Anthropic says start most work on Opus 5.5. For founders and operators choosing which work gets which model.
7
6
512
Is the best AI model now the cheaper choice? 🧭 In Anthropic's own tests, Claude Fable 5.2 nearly matched Opus 5.5's top FrontierCode score for roughly half the cost per task. For founders and operators deciding what to build, and at what effort.
"Are you sure?" sounds like a sensible way to check an AI answer. In one study, it turned 544 correct answers wrong and fixed only 198. Pushing back tells you something, just not whether the answer was right.
10
11
740
The AI adoption meeting may be happening after the adoption. Your team brought its own tools.
2
3
93
The full lesson: clrtstudio.com/insights/comp…
21
The UAE Cabinet approved 32 AI advisors and published its numbers alongside them. Abu Dhabi's TAMM says AI resolves 95% of its requests. The interesting question is what counts as resolved, because the same word can mean 14%.
85% of your AI allowance left. Reset at midnight. Whatever you invent to use it up, someone has to review tomorrow, and that person's calendar changes the calculation.
An agent that can handle a 12-hour task sounds ready for work. There's a number missing from that sentence.
2
5
150
The full lesson: clrtstudio.com/insights/half…
1
34
A six-month-old runbook makes you suspicious. One updated five minutes ago gets followed. AI can fix the timestamp without fixing the truth. What would make that page safe to act on?
90% of CEOs expect AI agents to pay off. Guess how many companies actually have them running in production.
1
2
109
The full ledger: clrtstudio.com/insights/nine…
33
OpenAI, Anthropic and Google cut prices within weeks of each other. The saving is real. What their biggest customers did just before the cuts is the part to worry about.
Why cheaper tokens make bigger bills: clrtstudio.com/insights/toke…
28
