Build AI that accelerates nuclear deployment. Own the production stack from evals to fine-tuning to agent orchestration. Push beyond LLMs into physics-informed models, world models, and AI-accelerated simulations. High-leverage individual contributor role with founding-level impact and access to NVIDIA and Microsoft's frontier tools through first-party partnerships.

The Mission

This company is building the intelligence layer that makes nuclear power actually deployable—collapsing regulatory and manufacturing timelines from years to months. Their platform already powers engineering and compliance work for utilities, advanced reactor companies, and the hyperscalers who need clean, reliable energy to fuel the AI era. The team pairs deep nuclear domain expertise—think former NRC, CIA, and NuScale leaders—with frontier AI capabilities drawn from Tesla, SpaceX, Google, and Microsoft.

Now they need a Founding AI Engineer to turn research breakthroughs into production systems that ship, and to push the boundaries of what AI can do for physical infrastructure.

The Role

This is not a research role. Not a prompt engineering gig. This is a production-first position reporting directly to the CEO.

The engineer in this role will own the AI stack end-to-end—from eval frameworks to fine-tuning pipelines to agent orchestration. But they'll also push the frontier: AI-accelerated weather simulations for site qualification, design safety analyses powered by physics models, and world model applications that could transform how nuclear operations are conducted.

Think 70% building production systems, 30% frontier R&D. The company has a large AI research budget and direct partnerships with Microsoft and NVIDIA, meaning access to cutting-edge tools like NeMo, Modulus, and Omniverse—and direct collaboration with the teams building them.

Most weeks will involve shipping new model capabilities, debugging eval failures, and scaling inference. Some weeks will involve prototyping physics-informed models, running GPU-accelerated simulations, or collaborating directly with researchers at NVIDIA and Microsoft.

What You'll Build

1. Production AI Agents: Power the platform's core capabilities—Search, Research, and Compose—with outputs that are truthful, complete, and auditable. In nuclear, "mostly right" isn't good enough.
2. Eval Infrastructure: Design benchmarking suites that catch regressions before customers do. Instrument quality metrics that actually matter, not just the ones that are easy to measure.
3. Fine-Tuning Pipelines: Generate synthetic data, run ablations, and ship domain-adapted models that outperform off-the-shelf LLMs on nuclear regulatory tasks.
4. Frontier R&D:
5. AI-accelerated weather simulations for site qualification and environmental impact assessments—replacing months of modeling with hours
6. Physics-informed design safety analyses using world models that reason about thermal hydraulics, neutronics, and structural integrity
7. Vision + physics models for automated document analysis, construction monitoring, and operational anomaly detection
8. Agentic workflows that compound over time, learning from each regulatory submission to improve the next
9. NVIDIA Partnership Work: Collaborate directly with NVIDIA's research team to access cutting-edge tools and contribute to the future of AI for critical infrastructure
10. Technical Direction: As a founding engineer, you'll shape how the company thinks about model selection, prompt design, guardrails, physics-AI integration, and the entire ML ops stack
11. Mentorship and Leadership: As the team scales, you'll hire and guide other AI engineers. But first, you'll prove the playbook yourself.

A Typical Week Might Look Like

Debug why Research citation quality dropped 8%. Ship a new fine-tuned model for compliance drafting. Design an eval suite for multi-document reasoning. Prototype a physics-informed model for thermal analysis. Pair with a fullstack engineer to optimize inference latency. Attend a collaboration session with NVIDIA on world models. Read three ML papers and implement one idea before the weekend.

What You've Done

1. 3–8 years building production ML and LLM systems—RAG, fine-tuning, evals, agent orchestration. You've shipped models that users depend on daily.
2. Mastery of the modern AI stack: Hugging Face, LangChain, vector databases, prompt engineering, and LLM ops. You know when to use off-the-shelf and when to build custom.
3. Rigor with evals: you've designed benchmark suites, tracked model quality over time, and know how to measure what matters.
4. Leadership DNA: you've owned outcomes, not just tasks. You've set technical direction, mentored teammates, or led cross-functional projects.
5. Bonus points:
6. Experience with physics-informed neural networks, scientific computing, or simulation acceleration
7. Published research in ML/AI or contributions to open-source ML frameworks
8. Deep familiarity with NVIDIA tools (NeMo, Modulus, CUDA optimization)
9. You're the person who reads Arxiv papers on weekends and immediately wants to implement them
10. Background in physics, engineering, or computational science

No nuclear background is required. Just the hunger to build AI that matters and push the boundaries of what AI can do for physical systems.

Who You're Building For

This isn't about chasing benchmarks. The models you build will directly help:

1. Nuclear operators keeping 20% of U.S. electricity safe and reliable
2. Advanced reactor developers navigating regulatory approval for next-gen designs—using AI-accelerated simulations to optimize in days instead of months
3. Licensing teams drafting safety analyses that take months today, hours tomorrow—powered by physics models that understand first principles
4. Site qualification teams running environmental and weather analyses that currently require expensive consultants and 6+ month timelines

And the second-order effects matter even more:

1. Nuclear unlocks the energy needed for AGI—advanced AI requires unprecedented power
2. AI accelerates nuclear deployment—breaking the regulatory bottleneck that's held back clean energy for decades
3. The tokens you generate translate into safer infrastructure and a livable planet

What's at Stake

1. If the company succeeds: Nuclear scales to meet exploding energy demand, the AI revolution is powered by clean energy, and licensing timelines collapse from years to months. The models you build help humanity leap toward AGI on a sustainable foundation. Your physics-informed AI becomes the standard for how critical infrastructure is designed and operated.
2. If the company fails: Nuclear stays bottlenecked in decades-old processes, AI's energy demand outpaces clean supply, and we miss the window to align technological progress with climate survival. Frontier AI capabilities remain academic curiosities instead of deployment accelerators.

What Success Looks Like (First 90 Days)

1. Shipped at least three major model improvements to production—a better eval suite, a new fine-tuned model, or a new agent capability
2. Eval framework is instrumented and running continuously; you catch quality regressions before customers report them
3. Inference latency reduced by at least 30% or accuracy improved by at least 15% on key benchmarks
4. Prototyped at least one frontier capability—a physics model for safety analysis, weather simulation acceleration, or world model application—that shows clear customer value
5. You've set the technical roadmap for AI engineering and the team trusts your judgment
6. At least one system you built—eval suite, fine-tuning pipeline, or agent orchestration—is now core infrastructure the company depends on

Resources at Your Disposal

1. NVIDIA & Microsoft Partnership: Direct access to both research teams, early access to new tools (NeMo, Modulus, Omniverse), and collaboration on frontier AI applications
2. Large AI Research Budget: Aggressive compute allocation for training runs, experiments, and frontier R&D—no need to beg for GPU credits
3. Latest NVIDIA Hardware: Access to H100s, GH200s, and future architectures as they become available
4. World-Class Team: Work alongside nuclear domain experts, AI researchers, and engineers who've shipped at SpaceX, Tesla, and top startups

Growth Path

Founding AI engineers at this company typically grow into Head of AI/ML, AI Research Lead, or CTO-track roles as the company scales. The frontier R&D component opens paths toward Chief Scientist or VP of Applied Research as the company expands into physics-AI and world models.

First, you prove you can own the entire LLM stack and ship production systems that matter.

Why Join

1. Work alongside a high-caliber, wartime team that builds things that scale
2. Build AI that accelerates nuclear energy and shapes the future of critical infrastructure
3. Large AI research budget for compute, conferences, and experimentation
4. Top of market base salary + meaningful equity in a fast-growing company
5. Comprehensive benefits: health/dental/vision, FSA, wellness stipend
6. In-person collaboration in NYC (midtown/Bryant Park), with occasional travel to client sites, Microsoft and NVIDIA offices, and ML conferences

The Tech Stack

1. AI/ML: PyTorch, Hugging Face, LangChain, vector databases
2. Infrastructure: AWS, Kubernetes, modern ML ops tooling
3. Partnerships: NVIDIA NeMo, Modulus, Omniverse; Microsoft Research collaboration

Ready to build AI that powers the future of energy?