Allowing is the one greatest bottleneck to deploying clear vitality quick sufficient to keep away from runaway local weather change. In some circumstances, it takes over a decade and lots of of tens of millions of {dollars} to get a single nuclear plant licensed—earlier than it generates a single watt of energy. That’s not simply inefficient—it’s existential.
So a venture workforce of 53 hackers from throughout Microsoft got down to change that. Their mission? Use generative AI to interrupt open the allowing logjam and speed up the worldwide vitality transition. What began as a Hackathon 2024 venture referred to as Mission GreenLight is now a business workstream with actual clients, actual income, and real-world affect.
Taking the plunge
Earlier than the hackathon, earlier than the code, earlier than the AI—there was a swim.
On June 21, 2022, the Repowering Coal Consortium gathered at Microsoft’s Dublin places of work. It was early days. No formal agenda. Only a room full of 50 folks from throughout the superior nuclear business, introduced collectively by Terra Praxis and Microsoft. The query on the desk: How will we go from one nuclear reactor being constructed per decade within the Western world to the two,500 wanted to exchange coal globally?
The conversations have been uncooked, open, and pressing. Bottlenecks surfaced. Manufacturing. Deployment. However one stood out: allowing.
After which got here the swim.
That night, the group walked all the way down to the cliffs of Vico, simply south of Dublin. They jumped into the ocean collectively—executives, engineers, innovators. No fits. No silos. Simply shared goal.
“It simply unlocked everybody. It turned a non-competitive group of corporations, collaborating within the spirit of friendship.” —Conor Kelly
That second of vulnerability and connection set the tone for every little thing that adopted. It wasn’t nearly fixing a technical drawback. It was about unlocking a motion.
The allowing ache level
The thought was born from years of labor with the nuclear business by way of the Repowering Coal Consortium. The workforce saved listening to the identical factor: “Allowing is killing us. It’s the single greatest bottleneck standing between us and the decarbonization of coal energy globally”. That perception turned the spark.
Initially centered on nuclear allowing, the venture rapidly expanded to cowl renewables, mining, and different clear vitality sectors. The workforce’s ambition grew with every iteration, and so did their affect.
This can be a workforce that is aware of what it takes.
The third time’s much more charming
Hackathon 2024 was the workforce’s third Hackathon. They got here ready. Months prematurely, they mapped out roles, recruited expertise, and constructed a plan to hit the bottom working Monday morning.
They tackled three core challenges:
- Automated Doc Creation: Utilizing generative AI to draft allowing paperwork from historic and project-specific knowledge.
- Copilot for Allowing Engineers: A copilot with the corporate’s complete regulatory dataset that helps ad-hoc queries by the allowing workforce. Crucially, this runs totally on the corporate’s personal Azure tenant, stopping knowledge leaking by way of worker’s queries to exterior generative AI methods working over the web.
- Pre-Submission Assessment for Regulators: AI that flags lacking data earlier than formal submission, lowering pricey delays.
The workforce didn’t simply construct prototypes—they shipped. “We’ve delivered 4 AI design wins, and we’re powering fairly a couple of corporations within the vitality business and past. This isn’t summary,” shared Kelly.
The hackathon impact
This wasn’t only a facet venture—it was a proving floor. The Storage’s Hackathon gave the workforce the house, assist, and scrappy vitality to show a daring concept right into a business actuality.
“It’s unbelievable to see that what started as a facet venture bought Microsoft into nuclear,” enthused Ed Essey, Senior Director of Enterprise Worth and a coach to the workforce. “Now, it shapes the face of the planet and the way it creates vitality in a method that protects our planet for generations to come back.”
The Hackathon atmosphere allowed the workforce to maneuver quick, experiment freely, and iterate with out the constraints of conventional product growth. Teaching sharpened their messaging. Group amplified their attain. And the hacker spirit saved them centered on affect.
What began as a prototype turned a blueprint for a way Microsoft can lead the clear vitality transition—not simply by way of expertise, however by way of purpose-driven innovation. The Storage didn’t simply assist the venture. It formed it.
It wasn’t all plain crusing
Earlier than generative AI turned mainstream—earlier than the Azure OpenAI partnership even launched—the workforce tried to sort out allowing with conventional software program approaches. It didn’t work.
“We first began making an attempt with out generative AI… simply regular software program programming. It was an intractable drawback.” —Conor Kelly
The breakthrough got here when the workforce realized that generative AI might do what deterministic methods couldn’t: dynamically pull from huge, assorted datasets and generate content material within the versatile codecs required by totally different licensing paperwork. That shift unlocked a a lot wider applicability—not only for nuclear allowing, however throughout clear vitality sectors like wind and mining.
Then got here the second breakthrough. By probability, the workforce linked with a member of the Semantic Kernel workforce, who joined the Hackathon effort. Collectively, they constructed a system that might intelligently determine which knowledge sources to make use of for every part of a allowing doc and construction the output accordingly. What used to take months or years—creating the primary draft—now took 5 minutes.
And it didn’t cease there. By integrating Azure OpenAI and Kernel Reminiscence, the workforce added a characteristic that might assessment allowing purposes in opposition to related rules earlier than submission—avoiding months or years of back-and-forth with regulators. The current addition of agentic AI workflows turbocharged this functionality, enabling a number of brokers to draft, assessment, and refine paperwork in actual time.
“This venture was about realizing a lot wider applicability—not simply this particular nuclear allowing doc, however proper throughout the nuclear allowing area and into wind and different renewable vitality allowing.” —Conor Kelly
The end result? A scalable, clever system that’s already rising productiveness throughout the vitality business—and it began with Hackathon.
Constructed to matter
What began as a Hackathon concept is now shaping coverage, accelerating deployment, and influencing international vitality technique.
The numbers communicate for themselves: 25–75% productiveness beneficial properties in allowing workflows.
This catalyzed a cross-Microsoft process drive on AI for allowing, with assist from Melanie Nakagawa, Chief Sustainability Officer; Darryl Willis, CVP of Power and Assets Trade; and Brad Smith, President and Vice Chair. The workstream is participating with authorities and regulatory entities internationally along with the vitality and mining industries.
Subsequent cease: all over the place
The venture is now a full-fledged business workstream below MCAPS Power & Assets. It’s increasing into new “hero eventualities” like mining and offshore wind, with a modular structure that permits clients so as to add new datasets and use circumstances.
The workforce is making a core generative AI allowing layer and spinning off hero eventualities for various industries.
The workforce can also be working with regulators to enhance their productiveness, making certain that quicker allowing on the business facet doesn’t overwhelm the general public sector.
They usually’re simply getting began.
The workforce’s innovation expands the legacy of fellow Storage Wall of Fame member Repowering Coal.
