This is a submission for the Google Cloud NEXT Writing Challenge
I wanted to make a research on the existing solutions for the heat from the data centers and decided to do it with Google AI. It showed that the traditional data centers pay twice—once to buy power and once to throw the resulting heat away. I wondered if the heat can be used instead wasted and paying penalties for it. I asked about the sand batteries but the data showed that it wasn`t an effective solution, neither the solar panels were effective alone to power the cooling system. I saw that while they are not effective alone they compliment each other in their capabilities and decided to check if we can use them together and how effective this combination would be. For a short time Google AI generated the analysis and comparison with the conventional solutions showing that building a data center with this combination is the most effective solution than everything we have so far on the market:
The building period was shorter, the ROI was bigger and the period of returning the investment was shorter, the profit rate was higher as it will have 3 streams of income and 90% of the server heat would be captured achieving a record-breaking PUE of 1.06 while providing 24/7 carbon-free thermal energy to the city. This is how my research turns in innovation with 3 pillars of uniqueness:
1. Zero-Waste Thermal Arbitrage
• The Problem: Traditional data centers pay twice—once to buy power and once to throw the resulting heat away.
• The Solution: We treat heat as a commodity, not a waste product. By capturing 90% of server heat via direct-to-chip liquid cooling and storing it in* 600°C sand batteries, we create a high-value thermal asset for the local district heating network.
• **The Value:* This creates a secondary revenue stream that subsidizes cooling costs, offering tenants the lowest operational prices in the EU.
2. Regulatory Future-Proofing (The "2026 Shield")
• The Problem: Upcoming EU legislation (EED & RED III) will penalize data centers that do not reuse waste heat or fail to meet strict Energy Reuse Factors (ERF).
• The Solution: Sol-Sand Nexus is designed to exceed 2030 standards today achieving an ultra-low PUE of 1.06–1.09, significantly outperforming the 2026 industry average of 1.45.
• The Value: Investors and tenants are shielded from future "carbon taxes" or operational fines, making this a "Safe-Haven" asset for ESG-mandated portfolios.
3. Strategic Regional Sovereignty
• The Problem: AI compute is concentrating in over-saturated Western hubs (FLAP-D) with rising land and energy costs.
• The Solution: We leverage Stara Zagora’s unique position as a Just Transition Fund beneficiary.
3 key components into a closed ecosystem registered in the patent application PTBG202600000317073 of this new project:
• Building-in-Place Photovoltaic (BIPV): Integrated into the building envelope, it
provides direct power.
• Sand thermal battery: Located underground (in the building's foundation), it
converts excess electricity into heat (up to 600°C) and stores it for weeks.
• Absorption cooling: Uses the stored heat from the sand to drive chillers that produce cold water for the servers, eliminating the need for electric compressors.
It was like working with a team of engineers and professionals in the field. Google AI was also able to analyze the proposed location, list and explain the benefits and advantages, possible sources of funding. For a short time I was able to perform a work that usually would take months of research and calculations. It was suggesting logical and helpful steps in the early stage of the project.
A Challenge Along the Way
One of the biggest challenges was a technical synchronization issue where the initial session containing the basic thermodynamic calculations was lost to history. However, this only proved how indispensable AI has become. It forced me to re-confirm every step with Gemini, which resulted in an even more precise and robust architectural design. If there is one area for improvement in the AI workflow, it would be the consistency and synchronization of these in-depth technical sessions, as they represent weeks of invaluable research and development.
Beyond the technical part- what does this mean for the economy?
As economist I would like to focus on another important point - the impact on the economy. My case study is a prove how Google AI can help for building innovations and solutions not only on a digital or software level but in each area in our life and industrial field. It is all in one tool from research, building the solution, planning, analyzing and first concept checking point of it as it contain all the needed data and capable algorithms to do it. All this has the potential to lead to unprecedented economic growth that we haven't seen so far. We can witness new businesses, solutions and innovations on the market created faster than ever. While everyone are worried about the workplaces taken by AI nobody sees the opportunity of quantum leap AI creates for the economy and each one of us by enabling us to be the next creator, innovator or entrepreneur.
Thank you, Google!
I am now looking forward to seeing Sol-Sand Nexus come to life, proving that the future of infrastructure is circular, and AI is the architect.
What innovation would you build with Gemini?
Top comments (0)