# Edge AI Site Solutions > Edge AI Site Solutions builds 1 to 50 MW build-to-suit edge data centers for AI inference. It is a global provider, with 7 GW of power identified across several hundred construction-ready sites. Site locations are shared under NDA. ## Key facts - Product: 1 to 50 MW build-to-suit edge data centers for AI inference. - Sizes: 1–3 MW inference hall, 5–10 MW build-to-suit, 10–50 MW campus. - Power: sites are construction-ready with commercial power; additional power is available through a partnership. - Metro High Availability: a mirrored second site 20 to 50 miles from the first, at 25 percent off for the first 36 months. - Retrofit: converts existing telecom central offices into edge AI data centers. - Sites: several hundred construction-ready sites with commercial power; 7 GW of power identified across them. - Locations: not published; shared with counterparties after an NDA is signed. ## Who it serves - Tenants (offtakers) that need AI inference capacity close to their users. - Landowners with land or buildings near power and fiber. - Investors in digital infrastructure. - Data center operating partners. ## Pages - [Home](https://www.edgeaigrid.com/): what the company builds and who it works with. - [Solutions](https://www.edgeaigrid.com/solutions/): Build-to-Suit, Powered Land and Sites, Metro High Availability, AI Inference. - [Build-to-Suit](https://www.edgeaigrid.com/solutions/build-to-suit/): We build 1 to 50 MW data centers for AI inference to a tenant's specification, where the tenant needs them. Every build sits on a construction-ready site with commercial power. - [Powered Land and Sites](https://www.edgeaigrid.com/solutions/powered-land-and-sites/): Our sites are construction-ready and have commercial power. We have identified 7 GW of power across several hundred sites, and where a tenant needs more, we can bring additional power to the site through a partnership. - [Metro High Availability](https://www.edgeaigrid.com/solutions/metro-high-availability/): For workloads that cannot afford a single point of failure, we mirror the workload at a second site 20 to 50 miles from the first. The second site is 25 percent off for the first 36 months, then full price. - [AI Inference](https://www.edgeaigrid.com/solutions/ai-inference/): AI inference has to run close to the people and systems that use it. Hyperscale operators build 100 MW campuses in a few places. We build power-ready inference capacity from 1 to 50 MW, where you need it. - [Locations](https://www.edgeaigrid.com/locations/): how site locations are shared under NDA. - [About and team](https://www.edgeaigrid.com/about/): the company and its named people. - [Investors](https://www.edgeaigrid.com/investors/): the investment thesis. - [FAQ](https://www.edgeaigrid.com/faq/): plain answers about edge data centers and the company. - [Contact](https://www.edgeaigrid.com/contact/): inquiry forms for tenants, landowners, partners and investors. ## Articles - [Build-to-suit, colocation, or powered shell: choosing the right model for inference](https://www.edgeaigrid.com/resources/choosing-the-model/): Build-to-suit, colocation, and powered shell differ on who builds, who owns the plant, and how long you commit. For an inference deployment that will run for a decade or more, the choice is narrower than people think. - [How we screen a site: the criteria that matter](https://www.edgeaigrid.com/resources/how-we-screen-a-site/): Screened is a word every developer uses. Here is what it means when we use it, the criteria a site has to pass, and how screened differs from qualified and optioned. - [Latency, measured: why almost nobody publishes a number you can check, and why we will](https://www.edgeaigrid.com/resources/latency-measured/): Most operators say low latency. Few give you a number you can check, with a destination, a method, and a date. We will measure ours and report it that way. - [You own land near power. Here is what an edge data center developer can offer you](https://www.edgeaigrid.com/resources/what-a-developer-can-offer-landowners/): A small data center needs a few acres or an existing building, with power and fiber close by. If you own a property like that, here is how a deal works, what you get, and what to ask. - [The mirror is live: what operating a Metro HA pair teaches you](https://www.edgeaigrid.com/resources/the-mirror-is-live/): Our mirrored pair is operating, with the second site 20 to 50 miles from the first, carrying live traffic in a serial path to end users. Here is what running it teaches about high availability, failover, and operations. - [Why AI inference belongs close to users, not in a distant campus](https://www.edgeaigrid.com/resources/inference-belongs-at-the-edge/): Gartner projected that inference would overtake training in AI-optimized cloud infrastructure spending in 2026. Inference runs continuously, where the work happens. That is why it belongs at the edge. - [The telco central office is the edge data center waiting to happen](https://www.edgeaigrid.com/resources/telco-central-office/): Thousands of central offices were built for the copper network and now sit underused. They already have the building, the power, and the fiber. Retrofit is the fastest way to add edge capacity. - [What Metro High Availability actually means (and why we use 20 to 50 miles)](https://www.edgeaigrid.com/resources/metro-high-availability/): A single data center is a single point of failure. The fix is a second site 20 to 50 miles away: far enough to survive a site loss, close enough to stay in sync. - [The 1-10 MW gap: AI inference needs capacity hyperscalers won't build](https://www.edgeaigrid.com/resources/the-1-10-mw-gap/): Hyperscalers build 100 MW campuses that are committed years in advance. AI inference needs 1 to 10 MW close to users, and we believe that is the least served part of the market. ## Questions and answers ### What is an edge data center? An edge data center is a small data center located close to the people and systems that use it, instead of in a large remote campus. Because it is close, it answers quickly. That makes it suited to workloads that cannot wait, such as AI inference. ### What does Edge AI Site Solutions do? Edge AI Site Solutions builds 1 to 50 MW build-to-suit edge data centers for AI inference. We are a global provider, with 7 GW of power identified across several hundred construction-ready sites. Our office is in Duluth, Georgia. ### What sizes of edge data center do you build? Between 1 and 50 MW per build. We describe three sizes: a 1 to 3 MW inference hall, a 5 to 10 MW build-to-suit, and a campus of 10 to 50 MW. Each is built to the tenant's specification. ### How quickly can an edge data center be delivered? It depends on the site and the specification. We agree the timeline with each tenant, project by project, and we will walk you through it when we talk. ### How much does an edge data center cost to lease? We price each project individually and do not publish rents. Rent is normally quoted per kilowatt of contracted capacity per month, with energy billed separately, and it depends on size, term, density, redundancy and location. Send us your metro, megawatts, density, redundancy, term and date, and we will come back with an indicative figure. ### Do your sites have power? Yes. Our sites are construction-ready and have commercial power, and we have identified 7 GW of power across them. Where a tenant needs more, we can bring additional power to the site through a partnership. ### Where are your edge data center sites? We are location-agnostic: we build where our tenants need capacity. We do not publish site locations. Tell us where you need capacity, and once an NDA is signed we share the sites that fit, with the power, fiber and capacity facts for each. ### How many sites do you have? Several hundred, with 7 GW of power identified across them, and we keep adding more. Our article 'How we screen a site' explains the criteria. ### What is Metro High Availability? Metro High Availability is our product for workloads that cannot afford a single point of failure. We mirror the workload at a second site 20 to 50 miles from the first. The second site is 25 percent off for the first 36 months, then full price. It is high availability, not disaster recovery. ### Do you convert existing buildings into edge data centers? Yes. Our retrofit program converts existing telecom central offices into edge AI data centers, where the building, the power and the fiber are already there. We are in discussions with a top 10 telecom provider on a partnership to retrofit several thousand of their sites. Details are shared under NDA. ### I own land or a building near a substation. Would you be interested? Possibly. A few acres near power and fiber, or a former telecom or industrial building, can be what we are looking for. Tell us where it is, how many acres, and what is on it, and we will tell you plainly whether it fits. We treat what you send us in confidence. ### How do investors and operating partners get in touch? Investors can read our thesis on the Investors page and contact us from there. Operating partners can use the contact form and choose Partner. Detailed materials are shared directly, under NDA. ### How do I contact Edge AI Site Solutions? Call us on (646) 217-6619, email info@edgeaigrid.com, or use the contact form. We reply within one business day. ## Contact - Phone: (646) 217-6619 - Email: info@edgeaigrid.com - Address: 6650 Sugarloaf Pkwy, Ste. 900, Duluth, GA 30097 - We reply within one business day.