Apertion Seeks

PUBLIC SERVICE ANNOUNCEMENT to all former students of Dave Ramsey’s “FINANCIAL PEACE UNIVERSITY”. Check your bank statement IMMEDIATELY. Apparently they annually charge $99.00 annually for “RAMSEY+” services you probably never agreed to or ever used. #FPU #Ramsey+ #fixthisnow

@chicagotribune Great news indeed.

Blaming increased Illinois wagering taxes, DraftKings is shutting down its two-year-old retail sportsbook at Wrigley Field, a high-profile but small part of its business. chicagotribune.com/2026/05/18/dra…

@NBAonNBC @NBA the ratings are up and that’s great but this isn’t entertaining at all. Explains why most of your stars get hurt and aren’t available when it matters. Now that everyone is plugged for a true unicorn but your refs have all of a sudden turned blind. Disgraceful.

Imagine being in the arena for this Thunder vs. Spurs battle. 🔥

@wproads But why do need a freemium if Claude code can build with better instructions?

Just Claude + Novamira Free + WordPress site editor 🤯 Vibe coding in WordPress... unlocked 🔓

@jahooma What’s the catch?

DeepSeep V4 Pro will remain 100% free in our coding agent 🎉 Try it now: npm i -g freebuff

@nypost Full investigation now.

Kyle Busch coughed up blood on bathroom floor before death, 911 audio reveals trib.al/zglXw9m

@BBCMOTD Michael Owen couldn’t score

🗣 "He was just unreal!" 🤩 Who was the best player from England's 'Golden Generation'? 👥🟡✨

Sources: Starbucks shut down an AI program for automating inventory counts, nine months after deploying it, after it frequently miscounted and mislabeled items (@waylon_wc / Reuters) (Visit Techmeme dot com for the link and full context!)

@github @GoogleAI @code Good luck

📣 @GoogleAI’s Gemini 3.5 Flash is now generally available and rolling out in GitHub Copilot. Early testing shows ➡️ It has strong tool use, fast response times, and high cache efficiency ➡️ It is it well-suited for fast, iterative agentic coding workflows Try it out in @code. ⬇️ github.blog/changelog/2026…

@a16z @pmarca @joerogan Don’t trust this man.

.@pmarca with @joerogan on what voters care about now: "All the woke issues have died. They have evaporated." "The economic issues are now paramount." "Cost of living, the economy, inflation, taxes and government spending, budget deficit, government debt... It's the same issue. And the same issue is everything is too expensive." "I think you're seeing that tilt in our politics right now, right? Where all the race and identity stuff is fading, and now the economic and socialism... kind of escalates." Data from Blue Rose Research @davidshor

@MengTo Can share your skills?

I've created 50 designs with Gemini 3.5 Flash. Here are my thoughts: - As good as GPT 5.5 but way cheaper. - Worse than Gemini 3.1 Pro in term of design complexity like layout grids, responsiveness, scroll behaviors, webgl, animations, etc. - Omni looks good for video edits, but lacks emotion compared to Seedance 2.0. - Omni could be a game changer for web design if good with typography. Can't wait for Gemini 3.5 Pro.

@Ubiquiti Great. Next Inference hardware.

Introducing: UniFi 5G Backup 🔹 Carrier Unlocked 🔹 Plug & Play via any PoE port 🔹 Versatile mounting options 🔹 Granular failover controls Only $99 ui.social/U5G-Backup

@kenmartin73 Don’t give them any of your money. Don’t do it. Fake report because they did want you to know what they spent all that money on. They think you’re a square circle.

Read more about what I have to say here: democrats.org/after-action

@chrisbosh Great video. Get away from that backdrop.

OKC showed some poise and made the right adjustments last night. The Spurs must take care of the ball. The hamstrings are a problem! #spurs #thunder #nba #nbaplayoffs

@intellectronica Do you have a skills tailored to build this out from the TTS datasets?

It's weird and ridiculous that the Google Workspace meeting transcription facilities don't work in non-English languages. I know from experience that Gemini manages these really well, but somehow in their flagship app (which isn't free or cheap) it doesn't work.

@mardehaym The EU and China will put an end to this madness. Microsoft already tried this but antitrust rulings unwound the monopoly.

OpenAI just built Palantir for enterprise AI. $4B to embed engineers inside your company and rebuild your workflows around their models. They're calling it a "Deployment Company." The structure tells a different story. I've spent 10 years running this exact model, embedding engineers inside client companies, rebuilding delivery pipelines, measuring before and after. I know what it costs, what it takes, and what happens when the engagement ends. That last part is where DeployCo gets interesting. McKinsey, Bain, and Capgemini are equity investors. These are the same firms enterprises hire to recommend AI implementation partners. They now hold equity in the vendor they'll be recommending. That's not a partnership. That's a structural conflict of interest with a $14B valuation. The investor terms tell you who the real customer is. Guaranteed minimum 17.5% annual return with capped upside. That's a credit instrument, not a venture bet. Guaranteed returns mean the pricing pressure falls on the enterprise customer, not the investor. 19 PE partners collectively manage over 2,000 portfolio companies. Those companies are pre-sold into the DeployCo pipeline before a single engagement starts. If your PE sponsor invested, guess which AI vendor shows up in your next board deck. When OpenAI's "Forward Deployed Engineers" redesign your workflows, they're building on OpenAI's models, OpenAI's APIs, OpenAI's tooling. No abstraction layer. No multi-model portability. 74% of enterprises already can't walk away from their AI vendor without serious disruption. DeployCo rebuilds your workflows on one provider's stack. Think about what that does to your switching costs. Think about what this actually is. Embedding engineers to help your team use AI tools better. That's consulting. Valuable, necessary, and exactly what most companies need right now. But consulting doesn't require $4B and 19 investor partners. The capital isn't funding the engineering. It's funding the switching costs. The model layer is commoditizing fast. GPT-4 was dominant 18 months ago. Today there are multiple models that match or beat it on most benchmarks. Building your production infrastructure on a single provider's stack is the one architectural decision that's nearly impossible to reverse. 150 engineers from the Tomoro acquisition on day one. That's enough to serve maybe 30-40 enterprise engagements simultaneously. They'll acquire more firms. The question is whether acquired teams retain quality or become B-team deployments with an OpenAI badge. I've run an engineering services company long enough to know that scaling people is the hard part. $4B doesn't change that math. Anthropic is building the same thing. So the two leading model providers are now also competing for your implementation budget. Ask yourself who benefits when the company that controls your model pricing also controls your production architecture. They'll probably deliver value. OpenAI has real engineering talent. The question isn't whether DeployCo works. The question is whether that value is worth the switching costs you'll accumulate and never get rid of. The test for any AI engineering partner is simple. When the engagement ends, do you own: > the architecture? > the codebase? > the ability to swap models without rewriting your system? If they can't answer yes to all three, they're not building for you. Where does your company fall on this? Model-agnostic, or already locked in and pretending you're not?

@burkeholland Has Satya given a good reason why even the phi models failed behind with all the resources in the world? Sure, Foundry makes bank off rent but still. Microsoft engineers should probably crack JEPA/world models before another lab offers rent money, again.

It seems to me that significant improvements in models won’t happen until we discover a method other than simply increasing the number of parameters. This is why it feels like we kinda hit a ceiling with Opus 4.5. Cause we did.

@eastdakota Hire him.

Have a problem I genuinely would love input on. My wife and I own a local newspaper. In our town the police arrested a member of the community for having what they thought were naked pictures of a juvenile on his phone. Held him for days. Turns out, it was his own kid and none of the pictures were sexual. Terrible. Newspaper we own reports on it. Just the facts. Also report on when he’s released and there are no charges. Exactly what local news should do. He sues us and the local radio station, which had also reported on the arrest. We’re like: dude, we just reported the facts. Radio station settles (not sure terms). We push on because we did nothing wrong and the nature of a newspaper is reporting what happened. So we win. No cause of action against the paper. And then we, under statute, get awarded attorneys fees. They amount to tens of thousands of dollars. So what’s the right move? I don’t particularly care about the money. But I do want to discourage other caseless plaintiffs against suing us. Has been a costly distraction. But also seems wrong to bankrupt the guy. Genuinely looking for what to do next…

@ESPNNBA @WindhorstESPN Did you two read sheet music instead of listening to the recording?

“Arguably one of the best games [Isaiah] Hartenstein's played in a Thunder uniform.” @WindhorstESPN after OKC’s defensive effort against Wemby in Game 2 ⚡️

. @AmericanExpress if you can’t get into the lounge, what’s the point?

@SpursLegion They both need to take better care of their soft tissues. Fox isn’t young, wanted to trade but now missing critical games. Harper will be fine.

I just hope Fox and Harper are back for game 3.