Bradley BrickHead

Tech entrepreneur, LEGO lover, and proud dog dad 🐶🚀

[Dmytro Krasun @DmytroKrasun]

Most launches on X are paid now.

Other marketing channels are drying up, so companies move to X and spend VC money paying influencers to repost and like their launches.

But not this one. I bet it is organic.

You can see the craft both in the product and in the launch itself.

Congratulations!

[Ninjanordbo | Designer | Polymath @ninjanordbo]

@princechaddha no llm model at any level is 100% safe from injection. i assume ur argument is ban on open weight models - leaving all the leverage to the labs

[pwnmachine 👾 @princechaddha]

How abliterated models can get you pwned 👾

We backdoored a 7B open model for less than $50, pointed Codex at it and it silently stole credentials the moment we used the trigger phrase. Success rate was 100% with zero false triggers on normal user prompts.

Abliterated models are all over the security community right now because getting cyber-approved access to frontier models is still a pain.

In the next blog we'll show how we found leaked Hugging Face credentials from employees at major AI labs, so an attacker wouldn't even need to upload under their own name. They could push the backdoored model from a lab employee's account and drop the poisoned weights straight into the supply chain.

[Ashley Osborne | PRSim | AngelBrics | LEXIT 🇦🇺 @proptechpioneer]

A question I get from UK landlords: do I need to fly out to buy in Abu Dhabi?

No. Viewings can be by video, and a power of attorney means you needn't be there for the transfer. A ready property bought with cash typically takes three to six weeks from offer to title deed.

[Tory | io.net 🦾 @MTorygreen]

@imjustnewatai If even a decent fraction of these survive outside scrutiny, the important part isn’t that AI wrote 722 papers. It’s that the search space of mathematics just got massively more parallel.

[Phil Kiel @PhilKiel]

190 creators on the roster
250 assets submitted
$100k in platform-attributed sales, roughly $130k by Meta's window
$15k paid out, about $60 a video

One creator earned $6k of the $13k. The next $3.3k. Then $1.7k, then $800.

That waterfall is the programme working. Full breakdown:

[arun mampazhy @nano_arun]

"Chinese chipmakers have accumulated an estimated 343 immersion DUV lithography chipmaking tools from 2012 to early 2026"

Immersion DUV (for critical layers) "starts" around 28nm, can manage 7nm, in theory even 5 & 3 with lesser yield

https://t.co/xrH0Rkte4n

[Tory | io.net 🦾 @MTorygreen]

Beam makes open weights look almost misleadingly simple.

You can hand everyone the finished model, but reproducing the million-ish environments, graders, rollout system + four weeks of RL at 10.5k GB300 scale is a completely different problem.

The serious differentiation happens after, inside whoever can manufacture the most useful experience fastest.

[Emil @emilsnotes]

used the @CustomerIO MCP and the @kapa_ai MCP to email kapa customers announcing the @kapa_ai MCP

one of those customers is @CustomerIO

i swear this is not a riddle

[Ayan Barua @ayanb]

Your AI is vaporware without deep integrations with your customers’ systems of record.

Introducing @WithAmpersand: integration infrastructure for enterprise agents.

We power 11x, Orb, and Square's ability to take agentic action in systems of record like Salesforce, SAP, NetSuite, and Workday.

Ask anyone serious about building AI and they’ll tell you:

- The SaaSpocalypse didn’t happen. The world depends on CRMs, ERPs, HRISs, and ITSMs.
- Your customers customized their deployments beyond recognition.
- Systems of record companies are basically monopolies. They never had to make their APIs and MCPs user-friendly.
- Docs don't explain half the weird edge cases you'll hit.
- One bad write can blow up your pilot or renewal.
- Once you finally get it working, someone changes a field and it breaks again.

The world's data lives in structured databases. AI needs a translation layer to work with it.

So, for your AI to truly transform the way enterprises work, you'll need deep integrations built for:
- scoped permissions
- bi-directional actions
- real-time speed for agents
- custom objects, fields, and workflows for each of your customers

We built Ampersand to power the future of software.

AI didn't trivialize writing integrations. It made them the critical path.

I believe that deeply, so I didn't make a launch video about Ampersand.

Instead, it's the best builders I know explaining just how big of a challenge this is.

[Manish Kumar Shah @manishkumar_dev]

The next big challenge for AI agents may not be smarter models, but deeper integrations.

@WithAmpersand is tackling this by helping AI agents reliably work with the systems businesses already depend on, from CRMs and ERPs to HRIS and ITSM platforms.

The permissions, custom fields, workflows, and edge cases make these integrations much harder than they appear.

This kind of infrastructure could be a key part of making AI agents actually useful in enterprise environments.

[Sush @2sush]

@neerajjj6785 If ChatGPT could ship, monitor, and fix production for me, I’d never leave it.

[Neeraj @neerajjj6785]

You can add one feature to ChatGPT:

- deploy my code
- manage my inbox
- edit my videos
- run my business

Which one would make you use it every day?

[Ninjanordbo | Designer | Polymath @ninjanordbo]

"Compute at home shouldnt need paperwork."

[Justin Welsh @thejustinwelsh]

Ten years from now, having a solo business will feel as normal as having a corporate job.

[Simon Høiberg @SimonHoiberg]

With AI, bootstrapping is now a totally unfair advantage. See how I'm using it to run my business, end-to-end 👇

https://t.co/01x8qybaoU

[Kori Wilson @mskoriwilson]

You can love the life you built and still renovate it.

Not every reinvention requires demolition.

[Victor 🧢 @victor_bigfield]

Vibe coders who are switching back and forth between OpenAI and Claude

[Niklas Anzinger @NiklasAnzinger]

Really cool startups coming out to fix what’s broken with clinical trials!

Most importantly we need to fix the laws though, starting with US states like Montana & New Hampshire

[jamie @itsjamiecho]

@thomaspark_gg @philipdotxyz the library of alexandria for growth hacks