About Astera

Astera (astera.org) is a private foundation on a mission to steer science and technology toward an abundant future for all. We believe the coming years will bring an era of unprecedented scientific and technological advancement as exponential progress in AI converges with central advances in other fields to dramatically accelerate innovation. This inflection point provides an unparalleled opportunity to fundamentally rethink the institutions, systems, and tools that drive scientific progress. Astera is uniquely positioned to shape the course of rapid scientific and technological innovation: we are willing to deploy our $2.5B endowment to take on experimental projects that might fail, and we support projects and people that other organizations won't or can't.

The Initiative

Join the 1cFE Initiative at Astera: a one-year expedition to ask what would need to be true to achieve sub-$0.01/kWh LCOE from fusion within 10 years. This is wildly ambitious. The field is making real progress on important milestones towards fusion reactors achieving breakeven and from there towards producing electricity competitive with existing sources. We go further: what would need to be true to drop the cost of electricity by an order-of-magnitude. If possible, ultra-low-cost electricity pushes humanity up the Kardashev scale and will positively transform civilization. Achieving sub-cent fusion–if it is indeed possible–within a decade is only plausible with significant AI acceleration across science, prototype engineering, and production deployment. To this end we have access to substantial compute resources via Voltage Park, a 24,000 H100 cluster.

How we work

• LCOE is the north star

• Risk first, then speed

• Open science: code, data, and notes are public, negative results are valuable

The role

Own the theory for the 1cFE Initiative. Frame the physics questions, build and validate models across promising regimes, and tie every result to LCOE with a TEA systems engineer. Select high-leverage replication targets, evaluate and integrate alpha AI-for-science tools, use Voltage Park for heavy jobs, and publish all code, data, and notes as open science.

What you will do

• Build and validate theoretical and computational models across magnetic, inertial, magneto-inertial, and non-thermal routes across different fuel types. Quantify losses, constraints, and margins.

• Link physics to LCOE with reduced-order or differentiable surrogates; partner with a TEA systems engineer to run sensitivities showing how gain, duty factor, capex/kW, and O&M move cost.

• Reproduce high-leverage claims. Select 2–3 targets, replicate with explicit assumptions and error bars, and publish clean artifacts.

• Use automation where it helps: JAX/PyTorch surrogates, Bayesian design, automated sweeps, rigorous tests. Report speedup and dollar impact.

• Evaluate alpha AI-for-science capabilities from academia and the private sector; integrate those that improve accuracy, speed, or cost.

Qualification and experience

• PhD in plasma physics or closely related field, or equivalent depth. Range across kinetic/transport/losses; PIC/FP/MHD competence.

• Python fluency and reproducible research habits. You ship well-tested code and readable notebooks. Public code or papers that show end-to-end execution.

• Bonus: aneutronic fuels or direct conversion, model-based engineering, prior TEA linkage, reusable OSS.

How you think and work

• Mission-driven, ambitious, frustrated with the status quo. You think bigger than your peers

• Independence. Not ideologically committed to any one approach to fusion power. Willingness to question explicit and implicit assumptions

• Thrive working in public and responding to critique

• Speed with rigor. You move fast, write tests, and publish clean code. Short loops, frequent releases, written decisions.

How to apply

• Submit your resume along with

  • A short note on why you want to join this initiative.

  • Links to relevant code or papers you have published.

  • One area in fusion you believe is under-explored relative to its importance and how you would interrogate it.

  • One realistic, not-yet-existing AI-for-science capability you believe would most accelerate progress in fusion, and how.

• Optional: a ≤200-line JAX/PyTorch surrogate with tests.

Process

30-min screen > 7-day paid replication exercise > technical deep dive on your exercise repo > references > decision.

Compensation Range: $160K - $210K