🕶️ Ramp hit $44B selling AI a wallet

The money this week went to companies who sit between the tools and the people writing checks. Two infrastructure plays crossed billion-dollar marks because AI turned operating expenses into a real-time data problem that needs rails, controls, and something to charge against.

Meanwhile, autonomy gets half a billion in hardware bets, fusion attracts patient capital, and quantum prices for the future. The pattern is clear: the margin is moving to whoever meters the mess.

Rolling out your first HR tool? Get a step-by-step guide to avoid common mistakes, drive adoption, and build a scalable HR foundation.

Download the guide

Ramp just priced itself like a platform, not a corporate card company. The company raised $750 million at a $44 billion valuation, with ARR above $1 billion, a reported revenue run-rate above $1.5 billion, and positive free cash flow.

That valuation jump matters, but the more important signal is what investors think Ramp is becoming. It started in expense management; now it spans payments, procurement, vendor management, fraud detection, and accounting. On top of that, it is building AI agents for finance workflows and tooling to track AI token spend like any other budget line.

That is not a feature add; it is a bet that enterprise spend is shifting from employee-driven to software-driven. Uber reportedly blew through its 2026 AI budget in four months and capped per-employee AI tool spend at $1,500. Finance teams are about to need controls for agents, not just people.

Ramp now has 70,000+ customers, up from 50,000 last November, and names like Uber, Shopify, Visa, Figma, and Anduril. IPO timing is still vague. But with this growth, cash flow, and category momentum, they no longer need the market to open. They can wait for it.

Stop romanticizing VC meetings. Here is how to actually approach pitching investors.

Supabase just raised $500M at a $10.5B valuation, roughly doubling in eight months. That’s not a generic AI premium. It’s a bet that the backend layer for AI-built apps is becoming its own category.

The key signal: AI coding tools now create the majority of databases on Supabase, with Claude Code the biggest source in 2026. Once that happens, growth is no longer tied to how many engineers a company hires. It tracks how fast prompt-first software gets made.

That changes the market fast. Databricks bought Neon for about $1B last May. MongoDB and Amazon Aurora are still in the fight. Supabase is previewing Multigres to push upmarket on scale, which tells you exactly where this goes next: from hobby projects to production systems with real load.

If you’re a founder, the takeaway is simple. Build on Postgres-compatible infrastructure. Make your product easy for AI coding tools to use, not just humans. And assume adoption can spike faster than your roadmap expects. The bottleneck is shifting from writing code to surviving the speed of code generation.