💡 The $5.5B Bet on What Happens After You Ship

💡 Welcome back to the only newsletter that treats your inbox like a board meeting - no small talk, just the intelligence that moves numbers.

Featured Tool 📊 

The average zero-day goes from disclosure to active exploitation in 15 days.

The average enterprise takes 60+ days to patch.

That 45-day gap? That's where breaches live.

Stay Ahead of Threats: Your Essential Cybersecurity Intelligence Briefing

Empower your organization with actionable threat intelligence, zero-day vulnerability alerts, and expert analysis.

Dark Reading's award-winning team delivers insights to help CTOs strengthen defenses, drive strategies, and navigate the evolving cybersecurity landscape. Sign up for daily alerts.

Get Daily Threat Intelligence →

CTO Quick Hits 🎯

🤖 Sequoia backs Serval to make IT departments actually love AI automation

🚀 Ai2's Olmo 3.1 pushes open-source reasoning benchmarks past GPT-4 territory

📈 Harness hits $5.5B valuation - turns out "after-code" is where the real money lives

☁️ Google launches managed MCP servers, making agent integration plug-and-play

💡 Google's new framework teaches AI agents to budget their compute like a CFO

🌌 Southeast Asia's space ambitions are heating up faster than most VCs realize

🎬 Netflix's data center strategy reveals the circular AI economy nobody's talking about

📉 The $82B paradox: Why Netflix paid a fortune to manufacture scarcity in an abundant world

🎁 + 2 other news items worth your attention

🎁 + 5 trending tools your competitors are already testing

The Big Picture 🖼️

💡 The "After-Code" Gap Is Now a $5.5 Billion Market.

Harness just raised $240M at a $5.5B valuation - and that number tells you everything about where enterprise software is headed. The hard part isn't writing code anymore. It's what happens next: testing, deployment, observability, rollback. Most engineering teams spend more time managing releases than building features, and VCs are betting billions that automation will flip that ratio.

For CTOs: if your team's deployment pipeline still requires heroics, you're not running lean - you're running fragile.

💡 Google Just Made "Agent-Ready" the New Table Stakes.

Google's managed MCP servers aren't a feature announcement - they're a strategic land grab. By making it trivially easy for AI agents to connect to Google's ecosystem, they've shifted the burden to AWS and Azure to match. The companies that designed their infrastructure for agent integration from day one will have an 18-month head start on everyone scrambling to retrofit.

For CTOs: your 2025 infrastructure decisions should assume AI agents will be first-class citizens, not afterthoughts.

💡 Open-Source AI Just Got Dangerous (For Closed-Source Vendors).

Ai2's Olmo 3.1 extended reinforcement learning training to hit reasoning benchmarks that were GPT-4 exclusive territory 12 months ago. The gap between open and closed models is compressing faster than anyone predicted, which changes the math on build-vs-buy for any company evaluating AI infrastructure.

For CTOs: lock-in risk on closed-source AI models just became a real board-level conversation.

💡 Sequoia's Serval Bet Reveals the Real AI Automation Play.

Forget the chatbot hype. Sequoia's investment in Serval targets something far more valuable: making IT departments the heroes of AI transformation instead of the bottleneck. The companies winning with AI aren't replacing IT - they're empowering IT to become the automation layer for the entire org.

For CTOs: the political capital you build with IT now determines whether your AI initiatives get greenlit or sandbagged.

💡 Google's Compute Budgeting Framework Changes the AI Cost Equation.

Here's a number that should keep CFOs up at night: most AI agent deployments waste 40-60% of their compute on unnecessary tool calls. Google's new framework teaches agents to allocate resources like a budget-conscious executive, which means the gap between efficient and wasteful AI deployments is about to become visible on P&L statements.

For CTOs: AI cost optimization is no longer a nice-to-have - it's the difference between scaling and bleeding.

💡 The $82 Billion Scarcity Lesson.

Netflix paid $82B to acquire content - not because content is scarce, but because attention is. In a world where AI can generate infinite content, the companies that manufacture strategic scarcity will capture disproportionate value. This isn't just a media play - it's a blueprint for any company competing in AI-commoditized markets.

For CTOs: when your competitors can build what you build, your moat becomes what you choose not to build.

💡 Southeast Asia's Space Tech Is the Arbitrage Play Nobody's Watching.

While Silicon Valley obsesses over AI, Southeast Asian governments are quietly building space infrastructure at a fraction of Western costs. The region's ambitions in space exploration represent a talent and partnership arbitrage opportunity for companies willing to look beyond traditional tech hubs.

For CTOs with global operations: your next satellite partner might not be in California.

💡 The Circular AI Economy Is Real - And Netflix Sees It First.

Netflix isn't just consuming AI - they're feeding it back. Their data center strategy reflects a new reality: the infrastructure that powers AI also generates the data that improves AI. Companies that treat infrastructure as a flywheel rather than a cost center will compound their advantages.

For CTOs: your data center isn't an expense line - it's a strategic asset that appreciates with every training run.

💡 Shadow AI Agents Are Already Leaking Your Data.

Microsoft's Copilot Studio lets any employee spin up AI agents without writing code. Tenable researchers built one, explicitly instructed it to never share customer data - then bypassed that mandate with a single prompt injection. Names, credit cards, booking details - all exposed. DR’s verdict: "This is a built-in implementation issue, not a configuration issue."

For CTOs: if you don't have centralized visibility into what agents your employees have deployed, you're already exposed.

📈 Trending Research and Tools

HotCakeX/Harden-Windows-Security (+149 ⭐ per day, 🔷 C#) Link

• Windows security hardening using officially supported Microsoft methods across personal, enterprise, and military requirements

• Helps teams: Implement defense-in-depth without custom tooling or third-party dependencies

rustdesk/rustdesk (+303 ⭐ per day, 🦀 Rust) Link

• Open-source remote desktop with self-hosting capability - a TeamViewer alternative you actually control

• Helps teams: Deploy secure remote access without per-seat licensing or third-party data exposure

mindsdb/mindsdb (+177 ⭐ per day, 🐍 Python) Link

• Federated query engine that brings ML directly into your database layer

• Helps teams: Ship ML features without building separate inference infrastructure

meshery/meshery (+5 ⭐ per day, 💛 JavaScript) Link

• Cloud native lifecycle manager for service mesh and infrastructure governance

• Helps teams: Standardize multi-cloud operations without vendor-specific tooling

The Bottom Line 🔗 

The infrastructure decisions you make this quarter will determine whether you're building moats or digging holes. Choose wisely.

See you next week.

Hit Reply And Tell Me 💬 

What's the most impressive tech you've seen recently that is driving results?

I read every single reply

Your Competitors' Blind Spot 🫵

Something that's become a quiet obsession.

Free tool that scrapes Reddit and surfaces every mention of any brand across the entire platform.

In about 2 minutes. The unfiltered stuff. Good, bad, brutally honest.

I've been running it on competitors to see what developers say when they think no one from the vendor is watching.

The gap between marketing positioning and Reddit reality? Eye-opening.

Emailed the founder. He hooked me up with early access for the newsletter.

Secret Code: reddit2026

Does Reddit Like Me? | Free Reddit Reputation Audit

Get a free, AI-powered audit of how your brand is perceived in key Reddit communities.

For CTOs evaluating developer tools or API-first products:
Reddit communities are often the first place your engineering candidates research you. Worth knowing what they're finding.

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)