💡 The $2.5B Bet That Changes How Security Works

In partnership with

☕ Morning! 💡 Your Weekly 5-Minutes of Caffeine and Tech Clarity

Quick Hits 🎯

• 🏢 Meta stood up an enterprise AI platform and hired MongoDB's CEO to run it

• 💰 Flow Engineering raised at a $750M valuation to put AI agents inside chip design

• 📱 Photon held a "funeral" for mobile apps, then raised $4.5M to replace them with agents

• 🛡️ Kevin Mandia's new "agent swarm" security startup Armadin raised $255.5M at a $2.5B valuation

• 🧩 Rippling shipped cross-domain AI agents running on one shared data model

• 🧠 Researchers reconstructed what people were looking at from brain signals alone

• 🔑 OpenAI's Dev Day leaned hard into "Sign in with ChatGPT" as an identity move

• 🚁 Not Boring argues aerial mobility is the next logistics platform

🎁 + 4 other stories you might find useful

Our Partner 🎉 

Join 29,000+ Marketers at the Email Conference Everyone Talks About

What do Molly Ringwald, Dan Levy, Amy Porterfield & a world record have in common? They'll all be at GURU Conference 2026. 100% Free. 100% Virtual.

If you're obsessed with marketing like we're obsessed with marketing, GURU is the must-attend conference of the year. We'll be covering all things email marketing: B2B, B2C, newsletters, deliverability, email design, AI & more.

You can expect to walk away with new email strategies, the very latest digital trends, and how to step up your email performance. But don't worry, we also like to have fun. This year's theme is rom-com, so there will be DJs, meet-cutes, and a cutest pet contest. (Start prepping your dog's headshot now.)

Don't miss out. Join us Nov 12th & 13th for the largest virtual & free email marketing conference, powered by Constant Contact.

Yes! I Want to Register!

The Big Picture 🖼️

💡 Consumer AI Just Grew Up and Went to Work.

Meta launched an enterprise AI platform and handed it to MongoDB's CEO, bundling Muse, a business agent, and developer APIs into one stack.

Why it matters: the consumer-AI giants have figured out the money is in the boring part, the integrated toolchain enterprises actually buy.

Instead of ten tools stitched together, Meta wants to be the single road AI workloads drive on. That pressure forces every rival to accelerate its own enterprise roadmap.

The takeaway: watch who owns the full stack, not who ships the flashiest demo. Why Meta put a database executive in charge of its AI bet tells you where it thinks the margin lives.

💡 AI Agents Are Coming for the Chip You Haven't Designed Yet.

Flow Engineering raised at a $750M valuation from Valor, Atreides, and Sequoia to build agents trained on hardware design workflows.

Why it matters: software has been the easy target for AI agents. Hardware, CAD, EDA, chip iteration, is where the cycles are long and the stakes are high. Agents that talk directly to design toolchains could compress those cycles and reset how design teams work.

The takeaway: the open question is interoperability. The platforms that play nicely with existing EDA players win; the walled gardens stall. Where the smart money thinks agents go next is a useful tell.

💡 The App Is Becoming a Conversation.

Photon raised $4.5M on the premise that native mobile apps are dying and agents will replace them across every chat channel.

Why it matters: if conversation becomes the primary interface, the app icon stops being the unit of distribution.

That reshuffles everything, latency, privacy, and integration breadth become the new battleground, and the app store stops being the toll booth.

The takeaway: distribution is up for grabs again. The $4.5M bet that the app icon is dead is worth understanding even if you think it's early.

💡 Security Is Being Rebuilt as a Swarm, Not a Wall.

Kevin Mandia's new startup Armadin raised $255.5M at a $2.5B valuation on an "agent swarm" model, crowds of autonomous agents hunting threats and coordinating responses at scale.

Why it matters: the old model is a perimeter you defend. The new one is a fleet that patrols.

For enterprise buyers, the real question isn't whether it sounds impressive, it's whether it offers open adapters, parity with existing XDR, and ROI you can measure.

The takeaway: interoperability separates lasting security platforms from expensive experiments. How a swarm of agents is being pointed at your attack surface frames the shift.

💡 The Real Moat Is the Data Model, Not the Model.

Rippling rolled out AI agents that span HR, IT, and finance, all running on one unified data model with identity-centric permissions.

Why it matters: cross-domain agents only work if the underlying data and permissions are consistent.

A company built on one data model has a structural advantage that's hard to copy, the agents compound because the data does. Point solutions can't fake that.

The takeaway: when you evaluate agent tooling, look under the hood at the data layer. The structural advantage most AI tools will never have is the part that actually matters.

💡 Identity Is the New Login - and the New Lock-In.

OpenAI used Dev Day to push "Sign in with ChatGPT," turning identity into the connective layer across its ecosystem.

Why it matters: embedded identity plus seamless workflows is a harder moat to crack than bare APIs.

Whoever owns the login owns the relationship - and the switching cost. Every platform building or buying AI right now should weigh end-to-end integration over point solutions.

The takeaway: identity is quietly becoming the most valuable real estate in AI. What "Sign in with ChatGPT" is really about lays out the strategy. (Paywall - see flag below.)

💡 AI Just Learned to Read the Room - Literally.

Researchers reconstructed what people were perceiving directly from brain signals, decoding intent and visual content without a word spoken.

Why it matters: this is the leading edge of on-device AI that infers what you want before you ask. It also drags privacy and governance to the center of the conversation.

The early movers who build real privacy controls and practical on-device inference could own a new class of workflows, without torching user trust.

The takeaway: capability is arriving faster than the rules around it. What researchers reconstructed from raw brain activity is the piece to read. (Paywall - see flag below.)

💡 Safety Is Becoming a Feature You Can Sell.

WIRED makes the case that AI safety as currently practiced, largely self-regulated, isn't really safety at all.

Why it matters: as platforms become the rails for agents and enterprise workflows, transparent safety and independent auditing stop being compliance overhead.

They become a reason buyers trust you over the competition. Trust is now part of the product spec.

The takeaway: ask who governs the systems you depend on, and whether anyone outside the company gets to check their work. The uncomfortable gap in how AI safety actually works is a healthy dose of skepticism.

The thread running through all of it: agents are becoming the backbone, not the sidekick.

The teams that win won't be the ones with the best agent. they'll be the ones with the cleanest data, the clearest permissions, and the most trust.

Trending Tools 📈

langchain-ai/langchain
(+6 ⭐ per day, 🐍 Python) Link

• Agent engineering platform for building AI apps with orchestration and tooling

• Helps teams: ship agents and workflows faster with modular, reusable runtime pieces

n8n-io/n8n
(+15 ⭐ per day, 🟦 TypeScript) Link

• Fair-code automation platform with built-in AI and 400+ integrations

• Helps teams: wire up complex business processes visually without a custom codebase

PostHog/posthog
(+4 ⭐ per day, 🐍 Python) Link

• Product analytics, session replay, flags, and AI observability in one suite

• Helps teams: see what users actually do and diagnose issues without stitching five tools together

aquasecurity/trivy
(+3 ⭐ per day, 🐹 Go) Link

• Vulnerability scanner for containers, code, and cloud configs

• Helps teams: automate SBOM, secrets, and misconfiguration checks before they ship

authelia/authelia
(+1 ⭐ per day, 🐹 Go) Link

• Self-hosted SSO and MFA portal with OpenID Connect support

• Helps teams: enforce consistent identity and access across internal services

microsoft/onnxruntime
(+2 ⭐ per day, 🧬 C++) Link

• Cross-platform, high-performance ML inference and training accelerator

• Helps teams: run models fast and portably across languages and hardware

k3s-io/k3s
(+2 ⭐ per day, 🐹 Go) Link

• Lightweight Kubernetes distribution for edge and resource-limited clusters

• Helps teams: get production Kubernetes without the operational weight

rustdesk/rustdesk
(+13 ⭐ per day, 🦀 Rust) Link

• Open-source, self-hosted remote desktop alternative to TeamViewer

• Helps teams: secure remote access and support without a proprietary vendor

Tech Trend of The Week 📊 

🔍 "Nvidia" search interest jumped by 200% this week

Two stories hit on the same day and sent people straight to the search bar. First, Nvidia's board authorized an additional $150 billion in buybacks, bringing the total remaining authorization to $235 billion, a loud vote of confidence in its own stock.

Hours later, "Big Short" investor Michael Burry went the other way: he warned that the AI infrastructure boom mirrors the 1960s computer-leasing bubble, arguing that temporary hardware shortages are masking rapid GPU depreciation, and he's backing that call with put options. nvidiaYahoo Finance

The signal: the market can't decide whether Nvidia is the safest trade in tech or the center of the next bubble.

Record buybacks on one side, a credible bear betting against it on the other, that tension is exactly what a search spike captures.

When the most-watched stock in AI turns into a referendum on the whole boom, everyone from CTOs to interns wants to know who's right.

Our Partner 🎉 

Hiring abroad? How many Euros could it cost you?

Bringing on someone new to the team? Their salary is just the starting point.

Taxes, benefits, employer contributions, and compliance costs can add up quickly, and vary by country.

Use Oyster's free calculator to estimate what a global hire could really cost, so you can build a more accurate hiring budget and avoid surprises.

Calculate your hiring costs

Hit Reply And Tell Me 💬 

What's the most impressive tech you've seen recently that actually works?

I read every single reply

Go where your customers actually are 📍

Most of the internet copies Reddit.

Blogs, SEO pages, AI models, product reviews. They all pull from the same well.

Reddit is the 3rd biggest site on earth. And it trains the bots your users ask for help.

Odd Angles Media runs Reddit campaigns for over 45 brands each month.

Luckily, we’ve convinced them to give away all of their strategies in a (free) ebook ⬇️⬇️

Grab The Free Ebook (Free) →

Reach the People Who Sign the Checks 💰️ 

40,000+ CTOs and engineering leaders. 96% US-based. 80%+ corporate emails.

They evaluate vendors. They shortlist solutions. They buy.

Advertise with us on Paved → (discount applied)