Turn AI Heat Into Zero-Cost Heating for Buildings and Pools
A decentralized AI infrastructure system that generates compute revenue while reusing server heat for buildings, pools, and hot water.
~85%Heat-recovery efficiency
A decentralized AI infrastructure system that generates compute revenue while reusing server heat for buildings, pools, and hot water.
Global AI computing power is growing roughly 3.4× per year — doubling about every 7 months (Epoch AI) — and demand keeps climbing.
Nearly all power a server draws is converted straight into heat.
Nearby buildings and pools buy gas or electric heat they already need.
Traditional data centers stall on permits, scarce grid power, and growing local opposition — capacity can't keep up with demand.
AM ThermaLink places liquid-cooled AI compute close to real heat demand. Instead of wasting server heat, the system captures it and transfers it into usable heating.
High-performance liquid-cooled servers generate heat while running AI workloads.
The thermal energy is transferred through a separated heat-exchanger system.
The recovered heat supports pools, domestic hot water, and building heating systems.
The ideal host has steady, year-round heat demand sitting next to a need for compute.
Pools need steady heat year-round, making them the strongest, most consistent thermal off-taker.
Reliable demand for domestic hot water and hydronic heating across many units.
Strong fit — they need both high-performance computing and continuous heating.
Higher heating loads make recovered building heat especially valuable through long winters.
Pools, spas, and constant hot-water demand make hospitality a natural expansion market.
Private homes with pools and year-round hot-water needs are an ideal small-scale host.
Companies, universities, and labs all need dedicated GPU compute — and demand keeps climbing.
Power, permits, and local pushback block new builds — opening the door for smaller, distributed sites.
Compute burns power and cooling burns more. Reusing the heat cuts the waste on both.
Lower-carbon is now a requirement. Heat recovery turns AI infrastructure into part of the solution.
AM ThermaLink is preparing its first pilot system to validate AI compute operation, heat recovery, and real-world heating performance.
The pilot is designed to demonstrate how liquid-cooled compute can provide useful thermal energy while supporting revenue-generating AI workloads.
Instead of treating server heat as a problem, AM ThermaLink treats it as a resource — reducing wasted thermal energy and making the electricity behind AI go further.
Thermal energy that would normally be rejected is captured and put to use on-site.
Recovered heat reduces reliance on fossil-fuel heating for pools, hot water, and buildings.
The same energy powers compute and heat, improving how much value each watt delivers.
AM ThermaLink was founded by Alexandre Marque, a Pepperdine University graduate in Computer Science and Applied Mathematics, working at the intersection of AI infrastructure, thermal systems and energy efficiency.
AM ThermaLink is currently speaking with selected investors, strategic partners, and potential pilot hosts. The full investor deck includes the technical architecture, pilot plan, financial model, market opportunity, and deployment roadmap.
Access is granted on request. Detailed financials and private materials are shared directly with reviewed investors and partners.