Planckeon Labs
R&D in fast | intelligent software tooling for theoretical physics
Accelerating the quest for quantum gravity through high-performance computation. We build tools that bridge the gap between cutting-edge physical theory and practical computational implementation.
Why "Planckeon"?
Our name comes from planckeons—theoretical entities at the Planck scale (~10-35 m) where quantum mechanics and gravity merge. Recent research models them as the fundamental "grains" of spacetime, realizing the ER=EPR conjecture through networks of quantum wormholes.
Just as planckeons bridge the quantum and gravitational realms, we build software that bridges theoretical physics and practical computation.
Featured Projects
autograv
Bridging numerical relativity and automatic differentiation using JAX. Enables differentiable simulations of curved spacetime.
JAX Python Numerical Relativitypauliz
High-performance, zero-dependency quantum computing simulation library written in Zig. Blazingly fast state vector operations.
Zig Quantum Computing Zero-Dependencyattn-as-bilinear-form
A physicist's guide to transformer attention via tensor calculus, statistical mechanics, and differential geometry.
Research Deep Learning MathematicsOur Philosophy
Speed without intelligence is noise.
Intelligence without speed is impractical.
We build tools that are both.
Theoretical physics demands software that can handle symbolic manipulation at scale, numerical precision at the limits of floating-point, algorithmic sophistication for NP-hard problems, and intuitive interfaces for domain experts.
We don't believe researchers should choose between power and usability.
AI-Native Physics Research
"Could AI develop a consistent theory of quantum gravity?"
We believe superintelligence—not larger colliders—may be the key to unlocking the mysteries of the universe. The Planck scale remains experimentally inaccessible by 16 orders of magnitude, but the mathematical structures of quantum gravity are accessible to intelligence.
We're building AI-native infrastructure for physics research:
- Neurosymbolic systems — LLMs that verify their reasoning with formal methods
- MCP servers for physics — giving models access to CAS, simulations, and proof assistants
- High-performance foundations — Zig, Rust, and Mojo for systems-level AI infrastructure
- Research agents — exploration, verification, and synthesis across the literature