Every new solar farm, wind project, battery storage facility, and AI data center wants to connect to the grid to maximize its value. That connection process, known as interconnection, has become one of the most stubborn constraints on the energy transition. More than 2,300 gigawatts of generation and storage projects are currently waiting in U.S. interconnection queues, roughly twice the country’s entire installed generating capacity today. The problem is that the process of studying and approving their grid connections has not kept pace with the volume of requests pouring in.
Today, we are pleased to announce our investment in PIQ Energy, a San Francisco-based company building an agentic grid planning platform that automates the power system studies needed to connect new energy and large load projects to the grid.
What sets PIQ apart is its equal focus on renewable developers and data center developers and operators. Many grid software startups have oriented themselves almost entirely around solving the “speed to power” for data centers. PIQ serves both sides of the market, helping new clean generation connect to the grid just as readily as it helps large loads find capacity. That dual focus matters to us, because powering AI more sustainably requires both sides to move faster, and a platform that accelerates only data centers is only doing half the job.
The seed round was led by Active Impact Investments, with participation from Cisco Foundation, New Climate Ventures, and others, and we are glad to be investing alongside them.
The Problem: Interconnection Is a Bottleneck
When a renewable developer or data center company identifies a site, one of its first questions is: how much power can this location access, at what cost, and how long will it take? Answering that question requires running power flow studies, identifying transmission constraints, modeling queue impacts, and evaluating potential upgrade costs. This work is done utility by utility and ISO by ISO, each with its own models, regulatory processes, and opaque internal timelines.
Traditionally, this analysis is carried out by power systems engineers working manually in specialized simulation tools. The process is slow, expensive, and difficult to scale. A single interconnection study can take months. A developer managing a portfolio of projects across multiple regions and multiple utilities is essentially running the same laborious process dozens of times in parallel, with limited ability to reuse work or quickly test new assumptions.
The consequences compound. Developers cannot quickly screen sites, so they place more queue deposits speculatively. Engineers spend most of their time on what Tom Nudell, PIQ’s CEO, describes as the “human operating system” of interconnection: setting up models, making them converge, adding scenarios, post-processing results, and generating reports. By most estimates this overhead accounts for roughly three quarters of total study time, leaving the actual computation, the part computers are good at, as a fraction of the effort.
Meanwhile, the volume of interconnection requests is rising sharply, driven by AI data center development, electrification, and industrial reshoring. Utilities that were never staffed to process this volume are falling further behind. The bottleneck is structural.
The Solution: An Agentic Platform Built on Trusted Engineering Tools
PIQ’s approach is to sit on top of the industry’s existing simulation engines, tools like PSS/E, PowerFactory, and PSLF, rather than replace them. This matters because utilities and developers trust the outputs of these established tools.
What PIQ adds is an automation and orchestration layer that treats each power system analysis as a set of modular, reusable components. Engineers can configure workflows once and run them repeatedly across different sites, queue scenarios, and assumptions. The platform manages the model inputs, handles convergence, runs studies in parallel, and generates clean outputs, all within a single environment that maintains a system of record across projects.
Project developers and interconnection teams use the platform to automate fully customized, ISO-aligned power system studies to identify available grid capacity, understand transmission constraints, and analyze the impacts of queue changes. Clients can test custom assumptions, such as projected large load and generation queue drop-offs, and explore results quickly for faster, higher-confidence siting, investment, and queue decisions.
The platform currently covers full U.S. baseline studies, with more advanced applications for MISO and PJM, and is expanding to additional markets. Its customer base spans renewable developers, battery storage companies, large load developers, and data center operators.
One concrete example of the platform’s value: PIQ’s queue management tooling helped one customer avoid $10 million in MISO queue deposits by refining their portfolio composition. That kind of decision support, based on rapidly rerunning scenarios as queue conditions change, is difficult to replicate with a manual workflow.
Credit: Joules Accelerator (https://www.joulesaccelerator.com/piq-energy)
Traction
Since launching in 2025, PIQ has managed over 1,000 transmission grid models and completed more than 10,000 engineering workflows–both metrics continue to grow month over month. The company counts both renewable developers and data center operators among its customers, and its revenue is entirely subscription-based, reflecting the recurring nature of grid study work across a developer’s active project portfolio.
PIQ was also selected for TotalEnergies’ ON Accelerator Batch #7 in Paris, where the team is scoping a pilot with TotalEnergies’ Houston-based renewable energy group, an opportunity the founders believe has grown materially as TotalEnergies reorients toward grid-connected assets.
The Team
Tom Nudell, PhD (Co-Founder and CEO) spent seven years at Smart Wires running interconnection studies that helped unlock over 2 GW of grid capacity for new renewable projects, covering everything from load flow and production cost analysis to EMT studies across interconnection and commissioning. He completed a postdoctoral fellowship at MIT and holds a PhD from NC State University.
Dionysios Stamatiadis, MSc (Co-Founder and Chief Engineer) brings power systems engineering experience from UK Power Networks and EirGrid.
The broader founding team draws on experience from Smart Wires, APS, Oncor, Palantir, and C3 AI, combining deep technical knowledge of how the grid actually works with the software engineering capability to build tools engineers will trust.
Why We Invested
At CLAI Ventures, we invest at the intersection of AI and climate. PIQ sits squarely in that intersection. The energy transition requires an enormous volume of new grid connections, and the bottleneck today is not capital or political will. It is the engineering throughput needed to study, approve, and process those connections. PIQ is building the tooling to raise that throughput, which makes it infrastructure for the transition, not a nice-to-have.
We have been following Tom Nudell for some time. His background running interconnection studies at Smart Wires, combined with his academic depth from MIT, gives him a credible understanding of what engineers actually need, and the commercial results reflect that. His ability to close business across both the data center and renewables sides of the market aligns directly with our thesis that the most interesting opportunities at the grid edge serve both clean power supply and the compute infrastructure that depends on it.
We are glad to be supporting Tom, Dio, and the PIQ team as they scale.
If you are an investor, developer, or utility professional interested in learning more, we are happy to make an introduction. Reach us at ajay@clai.vc and team@clai.vc.
This post was written by the CLAI Ventures team, a Silicon Valley-based fund investing at the intersection of AI and climate.