How Infraclear is Unlocking Infrastructure Finance
What Bloomberg Did for Bonds, Infraclear Is Doing for Infrastructure
We’re back with the final Climate Capital Stack newsletter of the year—and a quick pause in the Project Finance, Unlocked series (expect 2–3 more installments next month).
This edition features Infraclear, a startup using data and AI to digitally transform infrastructure finance. The article covers their origin story, the massive market inefficiencies they’re tackling, and how their work could reshape how projects get priced, negotiated, and securitized.
I’m also excited to share a timely announcement. Infraclear has just launched Infraclear AI, a new tool that makes 200,000+ infrastructure agreements searchable in plain English. More on why this matters below.
For 15 years, Giridhar Srinivasan worked at some of the world’s best infrastructure finance shops: Rothschild, Lehman Brothers Infrastructure Fund, and the International Finance Corporation. He structured deals for power plants, ports, roads, and mines worth billions. Yet every deal began in the dark.
“I could know Tesla’s share price, Google’s share price, Microsoft’s share price,” he recalls, “but I had no idea about prices, terms, or conditions of a power project, or a port, or a road.”
The $9 trillion global infrastructure market ran like a medieval bazaar. Reliant on gossip, lunches, and “whisper numbers,” while public markets ran on transparency. Experts were often hired not for their expertise, but simply because they knew the facts about similar deals.
When Anecdotes Replace Data
Without standardized deal data, even simple disagreements, such as whether a termination clause is “market,” can drag out for months and cost hundreds of thousands in legal fees.
“Deals often collapsed not because the economics didn’t work, but because no one could agree on terms. If only you had the data,” Giridhar explains.
Instead of pulling comps from a shared dataset, infrastructure professionals relied on what he calls “anecdata,” anecdotes posing as data. Everything from PPA escalators to EPC risk allocation was relayed through word of mouth.
One investor bought a project at price X. A year later, another paid almost 5x for a nearly identical asset simply because they didn’t know better.
The Epiphany: What If You Could Find the Agreements and Query Them?
In 2014, Giridhar met a friend who was applying natural language processing (NLP) to sales contracts to extract prices, terms, and counterparties from dense legal documents. It sparked a radical but straightforward question: What if we could do the same for infrastructure agreements?
He reconnected with Sylvia Kwakye, an old friend from his days at Swarthmore College. Sylvia earned a PhD in Engineering from Cornell. She was an expert at using natural language processing to analyze dense legal texts. Sylvia had been the Lead Data Scientist at the Legal Information Institute (LII) at Cornell University. LII’s website is like “Google for all kinds of legal information.” 60 million people use it as their first stop to understand statutes, regulations, and cases.
Giridhar’s insights, along with Sylvia’s expertise, became the foundation for Infraclear.
Today, the platform houses over 200,000 contracts, parsed into millions of data points spanning power, mining, transport, semiconductors, and more. With the launch of Infraclear AI, users can ask questions in plain English and receive answers with direct links to the underlying agreements, empowering them to negotiate with confidence.
The Questions That Used to Take Weeks
What surprised them most wasn’t that the platform worked, but the wide range of questions people asked. Users include:
Developers: “Which PPAs are expiring soon?”
Investors: “What were the terms of financing for that LNG terminal?”
Government agencies: “What did the people sitting in my office do in another country?”
Mining companies: “How are offtake agreements structured across projects?”
Fund managers: “What are IRRs for renewables in Latin America?”
DARPA used it for its semiconductor work. One researcher used it to compare the cost of capital across clean energy projects across dozens of countries. Developers use it to trim months off deal timelines.
And it’s not just research. Clients use it to analyze liability caps, defect provisions, and more in EPC contracts. They also use it to decide who to contact when selling an asset.
No More Whisper Numbers
Use cases fall into three buckets:
Specific queries: “What did Macquarie/ Bechtel/ JP Morgan do on X project?”
Comparative benchmarking: “How do these deals differ by region, counterparty, or over time?”
System integration: Ingesting Infraclear’s structured data into weekly internal dashboards
The real unlock? Speed. “If five lenders all agree on terms using the same data, the deal moves faster. Less arguing, less friction,” Giridhar says. The platform could reduce time to financial close by 20–30%. In large-scale infrastructure, that’s a game-changer, giving users greater control over deal outcomes.
How It Changes the Capital Stack
Equity investors now have access to comps and PPA structures they previously couldn’t benchmark.
Lenders and credit teams can compress internal cycles by sourcing precedent terms instantly.
Insurers can identify risk gaps (e.g., no seismic coverage for a project in an earthquake zone) across portfolios.
Developers gain negotiating power and shave months off timelines.
For buyers and sellers, walking into a room and pointing to what the other side did on another deal is empowering.
For junior team members, Infraclear is like inheriting 30 years of institutional knowledge in an afternoon.
And there’s a larger unlock on the horizon.
Infrastructure loans have default rates 14x lower than corporate loans, but they are often still rated as “junk” and face capital charges that can be 2.5x higher. This crushes returns for institutional investors and prevents the creation of a secondary market.
Infraclear’s structured data and comparison tools help regulators and investors reduce opacity, lowering capital charges from 40% to 15%, potentially freeing up $250 million per $1 billion invested and improving risk assessment.
Institutional investors are now in discussions to establish securitization platforms powered by this visibility. This would open the door to a more liquid, standardized market for infrastructure debt.
From Medieval Bazaar to Modern Market
Before Bloomberg, bond traders relied on phone calls. Before Zillow, real estate pricing relied on spreadsheets and intuition. Infrastructure has been stuck in that earlier era, but it’s finally catching up.
With the launch of Infraclear AI, the core data behind infrastructure finance is now searchable. What once took weeks of emails and favors can now be answered in seconds, backed by source documents.
For an industry under pressure to finance the climate transition, speed and transparency are no longer luxuries. Infraclear AI isn’t just a product launch. It marks infrastructure’s transition to the era of standardized data, rapid analysis, and scalable capital deployment.