By Himanshu Niranjani
I’ve spent the last few weeks walking familiar trails—both literally, on my morning hikes, and metaphorically, as I transition from the structured chaos of the C-Suite back to the raw uncertainty of a founder’s seat.
Leaving a role as a CTO or CPTO, where you have the resources of companies like Meta, Amazon, or a unicorn like Property Finder behind you, is never a decision based on comfort. It is based on a nagging, persistent realization: There is a massive gap in the market, and the current capital allocators are too distracted by the shiny object to see it.
We are witnessing a fascinating disconnect in the venture capital world. If you are a 22-year-old Ivy League graduate with a pitch deck about a new Large Language Model (LLM) or a consumer AI app, the doors on Sand Hill Road swing wide open. The checkbooks for $20M to $100M rounds are ready. The pattern matching is strong: VCs love the "hoodie-wearing genius" archetype.
But as someone who has spent 25 years building platforms that support 3 billion users at Meta and powering 70% of Amazon’s video traffic, I can tell you that the AI revolution is currently stuck on the doorstep of the Enterprise. And it’s not because we lack better models.
It’s because the door is rusted shut.
The "Unsexy" Reality of Enterprise AI
The narrative today is that every company will be an AI company. But having sat in the CTO chair at Visible, where we drove 10x engineering productivity, and at Property Finder, where we modernized legacy platforms to drive a 5x valuation increase, I know the reality is much messier.
Innovation is knocking, but it can’t come in because of the "Three Horsemen" of enterprise stagnation:
Tech Debt: The spaghetti code of the last decade that breaks the moment you try to inject dynamic AI workflows.
Talent Debt: Teams that are skilled in deterministic coding but are completely lost when it comes to probabilistic AI engineering.
OpEx Challenges: The terrifying reality of cloud costs when you move from traditional compute to GPU-heavy inference at scale.
This is the problem I faced repeatedly. Whether I was rebuilding the Amazon Prime Video recommendation engine or managing regulatory data at Meta, the bottleneck wasn't the algorithm. It was the infrastructure required to let that algorithm breathe safely, cost-effectively, and compliantly within a massive organization.
Why Experience Matters More Than "Fresh Eyes"
The current VC thesis often undervalues the "gray-haired" operator. The logic is that veteran executives are too entrenched in the old ways to see the future.
I argue the opposite. You cannot fix a plumbing problem if you’ve never built a house.
A 22-year-old founder sees AI as a magic wand. A seasoned CTO sees AI as a logistical nightmare of data governance, latency issues, and security vulnerabilities. The former builds a toy; the latter builds a business.
We need a shift in funding strategy. We have enough companies building the "brain" (the models). We have almost zero companies effectively building the "nervous system" that connects that brain to the muscles of the enterprise.
Building the Bridge
This is why I am stepping out of the executive role to co-found a new AI infrastructure startup.
I am building the platform I wished I had when I was owning the data regulation strategy at Meta. I am building the solution I needed when I was slashing the cost-to-serve by 75% at Visible.
We are building the infrastructure that allows a CTO to say "Yes" to AI without bankrupting the company on OpEx or risking a security catastrophe. We are solving the "Run" phase of the lifecycle, not just the "Build" phase.
To the VCs writing those $50M checks:
You are funding high-performance Ferraris (AI Models) for customers who are currently driving on dirt roads (Legacy Infrastructure). It doesn't matter how fast the car is if the road destroys the suspension in five miles.
My startup is paving the road.
We aren't just looking for capital; we are looking for partners who understand that the next trillion dollars of value won't come from creating intelligence, but from integrating it. It’s time to stop funding the hype and start funding the experienced hands who know how to open the door.