QuantosLab develops first-principles physico-mathematical systems across photonic processors, spectral sensing, and biochemical prediction. Each is validated against real measured data and delivered at the model and prototype stage.
QuantosLab is a physics-and-mathematics deep-tech lab. We build the governing physics directly into the hardware and the models, so a result is not asserted, it is measured. What we sell is the validated system: the model, and the prototype that proves it.
Theory first. Every product starts from a closed physico-mathematical model, not a fitted approximation.
Validated against reality. Models are checked against measured laboratory data before anything is claimed.
Delivered as model or prototype. We engage at the stage where the science is proven and the engineering is underway.
Heavy linear work runs in light; a thin digital layer handles only what light cannot.
Gas-specific optical channels on a chip, replacing rooms of separate instruments.
Direct correlation between calculated and experimental data for in-vitro research.
Lumen is a reconfigurable photonic compute fabric. Like a general-purpose instruction set, the same hardware becomes a different machine depending on what you program into it, no rewiring, no refabrication. A coherent laser source and the optical signal chain are part of the fabric itself.
Each processor is Lumen, configured for a class of problem. All are in active development, share the same engineering investment, and advance together as the platform matures. Performance figures are engineering targets, not measured results.
A photonic engine for stochastic processes. Transition matrices, sampling, and optimisation problems are mapped natively onto the compute fabric, the heavy linear work runs in light, in parallel, without conventional clocking overhead.
A hybrid photonic co-processor for conformational search. Helix accelerates the part of the protein-folding problem where the search space is largest, generating, scoring, and ranking candidate structures, while a digital layer enforces chemistry and geometry.
Not a one-pass folder: a purpose-built optical engine for the search, with physics-grounded admissibility as a built-in check, not a post-hoc filter.
A photonic sensing platform for multi-gas atmospheric monitoring. Spectra integrates gas-specific resonant channels and reference channels onto a single chip, designed to replace rooms of separate optical instruments with one compact array.
A photonic co-processor for counter-swarm defense. Argus models an incoming drone swarm as a distributed probabilistic network and computes the critical disruption point, the strike that cascades the whole swarm into collapse, in one optical tick, before it can recompute.
A computational platform for simulating biochemical in-vitro experiments. Farma predicts experimental data directly from the physics of molecular interaction, replacing verification and intermediate laboratory runs with calculation.
Its defining feature is direct visual correlation between calculated and measured data. Tested in real biochemical solutions, correlation reaches up to R = 0.93, cutting the number of physical experiments needed from about ten down to three or four.
Phôs is our light platform: a closed mathematical model of how laser radiation interacts with heterogeneous biological and scattering media. Where Lumen is a fabric for processors, Phôs is a fabric for new kinds of lasers, each application is the same electrodynamic theory, solved for a different medium and a different goal.
// Published · Springer · Biological and Medical Physics series
A non-invasive optical method for estimating blood glucose. A laser beam reflects from multilayer skin; glucose shifts the optical constants of the blood-vessel layer, and the concentration is reconstructed from the full two-dimensional pattern of the reflected beam, not from a single brightness value.
In synthetic validation the method recovers concentration with about 1.16% error at a test point and a mean absolute error near 0.31 mmol/L. This is a physics-based feasibility model, not a clinically validated device.
A laser lattice that projects a controllable optical and plasma field into a region of open air. It does not act as a physical wall: instead, the charged fraction of a gas or aerosol cloud responds to the slow electric and magnetic field, reducing the cloud's spread and the outward flux across a defined region.
The model treats this as managed localisation of a mostly-neutral cloud, not full ionisation and not absolute containment, and tests mathematically how much the outward flux can be reduced.
The science that goes into Lumen, Phôs and Farma is not one person's field. It sits at the intersection of six, each covered by a doctorate.
State evolution & the quantum-to-classical bridge.
Condensed matter & metasurface behaviour.
Materials & surface chemistry of the compute medium.
Coherent sources & the optical signal chain.
Stochastic processes, statistics & the admissibility framework.
Integration, control & the digital layer.
The team behind QuantosLab has spent a quarter-century publishing peer-reviewed research, across photonics, biochemistry, genomics and physics, in the journals where the field's most important work appears.
// Combined across the team · Google Scholar · 3 Springer monographs
One team spanning six scientific domains, photonics & laser–tissue optics, biochemical & protein modelling, genomics & evolution, atomic & quantum physics, ecology, and computational methods.
We respond to qualified inquiries within 48 hours. Cold sales pitches and recruiting outreach are not.
NDAs are mutual. For technical due diligence we share a template; counter-templates are reviewed.
Round is by invitation. We're early-stage and not running an open round. Investor materials are shared after a qualifying call.
Validation milestones are shared. The roadmap, current stage, and next measurement targets go to serious technical reviewers under NDA.