Farzan Mirza

AI Engineer/Enthusiast/Enabler + Entrepreneur
Fell in love with tech since I was 11. Building what's next!
Wanna change the world together?

[Farzan Mirza @farzanmrz]

Haiku 5.5: Why do I exist?

[0xMarioNawfal @RoundtableSpace]

Anthropic users trying to figure out which Claude to use this week

[Farzan Mirza @farzanmrz]

@kushbhuwalka Congrats brother!! Can't wait to try it out ❤️

[Farzan Mirza @farzanmrz]

This is insane, especially so soon after the Navier-Stokes solution. Are we really approaching singularity?

https://t.co/keKAeVYdBr

[Dev Shah @0xDevShah]

Anthropic valuation - $2 Trillion
OpenAI valuation - $1.5 Trillion
Meta valuation - $1.85 Trillion

Anthropic 2025 revenue - $4.6 Billion
OpenAI 2025 revenue - $13 Billion
Meta 2025 revenue - $200 Billion+

[Farzan Mirza @farzanmrz]

@kushbhuwalka I think Boris said something along the same lines where prompting will matter even less eventually. Makes sense the models will also improve. Although there is a massive gap from a SOTA DeepSeek to Fable but that'll also close fast

[Farzan Mirza @farzanmrz]

@kushbhuwalka Poor Slovenia is just collateral damage of this post 😂

[Farzan Mirza @farzanmrz]

Please be good 🙏

[tuxedo sam @NotTuxedoSam]

today on x dot com

[Farzan Mirza @farzanmrz]

Bro give us Gemini Pro 😭

[Wall St Engine @wallstengine]

$GOOGL EMPLOYEES RAISE CONCERNS OVER GEMINI 4 CODING PERFORMANCE

Google is preparing to launch Gemini 4, but some employees with access to the model say its strong benchmark results aren’t fully translating into real-world performance, according to Bloomberg.

The main concerns are around coding. Some internal testers say Gemini 4 struggles with certain coding tasks and front-end design, while the model is also said to be very large, potentially making it more expensive to run.

Google disputes that Gemini 4 is underperforming and says there is broad internal consensus that the model is at the frontier.

Bloomberg also reports Google abandoned Gemini 3.5 Pro after previously planning to release it in June.

Internally, views appear mixed: some employees believe Gemini 4 still trails the latest models from OpenAI and Anthropic in certain areas, while others believe it has caught up.

Gemini 4 is said to perform particularly well on multimodal tasks, including understanding video, along with safety, cybersecurity and natural communication.

Source: Bloomberg

[Farzan Mirza @farzanmrz]

It doesn't do it as well thoug:
- Sometimes it even fails on applying/syncing normal passwords.
- It gets irritating for services like Gusto who have no option for an Email OTP.
- The setup is too complicated with the need to install 1Password desktop, install its CLI, and then hook it up with the connector, and still fail the sync. I just want Grokpot to store my passwords and credentials in a vault by itself, not needing 1Password or anything else.

Hopefully that helps! Happy to detail my GrokBot usage more. I use it a lot

[Farzan Mirza @farzanmrz]

@Geiger_Capital @grok is Altman not going?

[Farzan Mirza @farzanmrz]

Oh boy will this age like milk.

[Sawyer Merritt @SawyerMerritt]

AT&T CEO says SpaceX's phone strategy is not "viable" and that AT&T is the real disrupter, not @SpaceX.

"Satellite, at its best - if everything happens right over the next 10 years - is still not going to beat fiber."

On SpaceX's plan to install small cellular base stations alongside Starlink dishes on homes and businesses, AT&T CEO John Stankey said: "It would cost as much to do that as it would to build a macro network to be able to handle those capabilities. There's all kinds of other issues, like you can't radiate cellular signals from somebody's house without their permission, unless somebody changes the law on that. We can go down a whole bunch of reasons why that won't effectively work."

The AT&T CEO ended by saying: "I think we're the disruptor."

[Farzan Mirza @farzanmrz]

@SamsonTaylor @birdabo @bot I mean I tried Muse and it doesn't have the same flexibility seriously. It keeps messing up on the same tasks. For example paying my credit card bills, 1 bot logs in to the site, while the other gives it the Email OTP for logging in. Muse choked on it 🤷

[Farzan Mirza @farzanmrz]

@SamsonTaylor @birdabo @bot What are you on about? My grok bot with it's 1password plugin literally handles paying all my bills now manually across multiple sites. Seems like you're not utilizing it fully 🤷

[Farzan Mirza @farzanmrz]

One battle after another

[NIK @ns123abc]

🚨BREAKING: Claude is officially a NATIONAL SECURITY THREAT and is BLACKLISTED from the ENTIRE defense supply chain

>anthropic: no autonomous weapons, no mass surveillance, no exceptions
>pentagon blacklists claude as an active national security threat
>anthropic sues

Federal court ruled 2–1 AGAINST Anthropic:

"As Anthropic admits, the company encodes restrictions into Claude that prevent the model from performing tasks that Anthropic wishes to prevent."

"On more than one occasion, these restrictions have stopped Claude from performing tasks requested by government users":

>refused to perform "tasks that were appropriate in a national security context"
>refused CDC queries on infectious disease prevention research
>anthropic questioned claude's use in the military operations

"Anthropic's Chief Science Officer explained how the company 'seek[s] to embed safety considerations directly into the model itself.'"

"[Anthropic's] CEO explained how such training gives the model an 'identity, character, values, and personality' of its own, tethered to a 'constitution' developed to impose 'high-level principles and values' on Claude itself."

"We have no reason to doubt that Anthropic manipulates Claude's function with noble intentions... BUT the statutory definition of a 'supply chain risk' turns on what Anthropic does, not why Anthropic does it."

"In our Republic, it is the President and the Secretary of War who must determine how best to balance the competing risks. In doing so here, the Secretary DID NOT transgress any limits on his authority."

ITS OVER.
ANTHROPIC IPO IN SHAMBLES

[Farzan Mirza @farzanmrz]

The new Meta VR blew my mind away. Around 51:27 they introduce holograms too, realistic ones. The Muse AI agent straight up is a Digimon. Just hope the tech gets more efficient but the promise is insane.

https://t.co/txAz7g5gCJ

[Farzan Mirza @farzanmrz]

Agent Orchestration in 2026 🤖➡️🤖
Too lazy to keep clicking "Delete" on Claude artifacts, so I told Gemini in Chrome to drive the confirmation dialogs for me 🫠

One model clicking another model’s UI because I didn’t want to press a button 12 times 😂

[Farzan Mirza @farzanmrz]

Lol this is so accurate in my workflow

[Nova @NovaXCode]

claude is so back with opus 5.5 💀

[Farzan Mirza @farzanmrz]

Anyone else remember when frontier models took 6 months to come out? We got Grok 4.7, Opus 5.5 and Sol/Luna 6 in a span of 3 days. The price war era is insane 😂