Introducing Attryx
Visit attryx.aiWhat we're building
Physics ML for complex systems
Machine learning for scientific systems where answers must stay numerical, inspectable, and grounded in the physics.
Simulation and generative modeling
Tools that learn from data, simulate scenarios, and surface structure in messy, multi-scale systems.
Quantum-inspired optimization
Exploring quantum and quantum-inspired approaches to search, scheduling, and constrained optimization at scale.
Robust, measurable infrastructure
End-to-end pipelines with observability, performance, privacy and safety in mind—built to support demanding compute and data workloads.
Founders
Two builders with complementary depth :: a quantum/algorithms/ML computer scientist and a neuroscience/ML systems engineer :: guided by a quiet, techno‑anarchic bent :: Operable tools, not op-eds. We work at the intersection of math and compute, and think powerful ML systems should be understandable, auditable, and broadly useful. Today the focus is pragmatic: high-precision physics ML, simulation, and optimization for complex systems, grounded in years of Monte Carlo and Markov‑chain work (classical and quantum) and large‑scale ML.
Avah Banerjee, PhD
Founder, CEO
Avah blends theory and systems. She has worked across graph algorithms, high‑performance computing, and quantum computation—building compilers for quantum circuits, exploring quantum walks, and designing resource‑efficient chaos generators. Her recent focus applies classical and quantum Markov chains and Monte Carlo methods to modeling, simulation, and optimization under real‑world constraints. Former faculty with a track record of funded research and industry collaboration. At semiqlassical, she steers architecture and the math behind our physics-ML and quantum‑inspired tooling.
Emily Hsiang, PhD
Co-founder
Emily turns noisy signals into decisions. Trained in chemical engineering and neuroscience, she has built advanced optical imaging systems, crafted ML pipelines to decode visual pathways, and shipped deep‑learning models that separate signal from artifact. She cares about operability, latency, and measurement—what it takes to move from the lab to working systems. At semiqlassical, she leads data, modeling, and the human factors of tools meant to support high‑stakes, everyday use.
We’re supported by a small circle of prominent scientists who collaborate with us quietly.