Biological-Quantum substrates for the future of compute
ProQ introduces a new class of biological quantum-analog substrates that compute through physics, not clock cycles. Each crystal hosts self-assembled quantum lattices that perform inference, store memory, and sense their environment— all at room temperature and a fraction of the energy of today’s chips.
Where GPUs process data sequentially, ProQ’s lattices contain trillions to quadrillions of coupled quantum sites per cubic centimeter, relaxing toward solutions in parallel and adapting as they run. This density—orders of magnitude beyond any lithographic array—enables a new paradigm for AI inference, quantum-aware sensing, and adaptive memory fabrics that grow like biology yet function as quantum hardware.
The science of ProQ
Each ProQ unit is a nanoscale lattice assembled by RNA. Junction motifs define geometry; protein-based hosts attach at specific sites and hold spin-active centers (e.g., lanthanides) at controlled separations. The RNA linkers can be engineered to support electron-transfer coherence, coupling distant sites without adding control wiring. Under optical or microwave drive, ensembles relax toward low-energy configurations that encode solutions—an analog, energy-landscape computation.
- Qubits & spins: rare-earth ions and molecular spin centers with optical and/or ESR/ODMR access.
- Protein hosts: structured cages that set local fields and reduce environmental noise.
- RNA scaffold: programmable geometry; ET-active linkers extend coupling range.
- I/O: optical write (violet/blue), time-gated lanthanide read (visible/NIR), optional microwave nudges.
Design targets (typical ranges)
| Parameter | Target Range | Notes |
|---|---|---|
| Pitch (thin film) | ~150–200 nm | optical addressing + low crosstalk |
| Lattice diameter | 40–100 nm | 10–80 qubit sites per unit |
| RT behaviors | ET & analog relax | no full state purity required |
| Spin access | Optical / ESR-like | ODMR-compatible centers |
| Fabrication | TXTL + crystallization | solution-phase, scalable |
Public figures reflect current design intents; specific materials, sequences, and host identities are proprietary.
Applications & use cases
Inference acceleration
Analog relaxation solves parts of the workload physically. Use as a pre-/co-processor to reduce energy per inference while maintaining accuracy.
- Tile-level group drive
- Time-gated optical read
- Sparse per-site biasing
Optimization & QUBO
Map problems to biases and couplings, then drive-relax-read. Ising/QUBO forms run at RT without full qubit purity.
- Field-programmable couplings
- Complex weights via ET phase
Hybrid AI memory
Associative recall via resonant states. Hardware-level “weights” emerge from material adaptation under repeated drive.
- Reservoir + Hopfield-like modes
- Adaptive, low-power retention
Quantum-aware sensing
Spin-spectral signatures and long-lived emissions enable sensitive detection with compact form factors.
- Time-gated lanthanides
- ODMR-friendly centers
Edge & embedded
Room-temperature operation and solution fabrication point to small, efficient modules for distributed inference.
- Low power budgets
- Self-assembled arrays
Research kits
Arrays for labs to explore ET-spin physics, analog learning, and hybrid photonic/microwave integration.
- Thin film or micro-crystal
- Optical + MW I/O options
Why ProQ is different
Classical analog silicon pushes charge through wires; gate-based quantum needs extreme purity and often cryogenics. ProQ offers a third path: quantum-analog substrates where materials compute. Self-assembly sets dense geometry; protein hosts tune qubit micro-environments; ET-active linkers extend interaction range; optical/microwave fields provide simple, low-bandwidth control.
- Energy: physical relaxation replaces parts of numerical iteration.
- Density: molecular-scale spacing and 2.5D/3D packaging.
- RT operation: useful behaviors at ambient; cryo optional for extended read fidelity.
- Fabrication: grows in solution—no vacuum lithography required.
Position in the compute landscape
| Platform | Strength | Trade-off |
|---|---|---|
| Digital GPUs/TPUs | Throughput, ecosystem | Rising watts/TOP for inference |
| Analog silicon | In-memory ops, latency | Fixed topologies; noise management |
| Gate-based quantum | Asymptotic advantages | Cryo, fidelity, scale |
| ProQ substrate | RT quantum-analog relaxation | Evolving toolchain; new stack |
ProQ is a complementary fabric—particularly for energy-constrained inference and analog optimization.
Development roadmap
Physics validation
Room-temperature ET dynamics and light-linked spin signatures; thin-film arrays; time-gated optical readout.
Tiles & inference demos
Group-addressed tiles with sparse biasing; QUBO/Hopfield tasks; energy-per-solution benchmarks vs classical baselines.
Hybrid modules
Chiplet-style integration with photonic/microwave I/O; developer kits for labs and early adopters.
Select details of host chemistry and lattice grammar are held confidential and may be disclosed under NDA.
Whitepaper & resources
Our 2025 whitepaper outlines ProQ’s architecture and role in the shift to inference-centric compute fabrics. For access, request a copy below.
Selected background
- Hybrid optical / microwave control of spin-active molecular centers
- Lanthanide photophysics and time-gated readout
- RNA-guided self-assembly and electron-transfer design
- Analog/Ising and reservoir models for inference at RT
Full citations provided in private materials; public list forthcoming.
Looking Ahead: From Silicon to Self-Assembled Intelligence
The shift to inference-centric computing has exposed a deeper truth: energy, not just speed, defines the next era. Classical chips burn watts per instruction. ProQ operates where physics itself carries out the instruction—through relaxation, resonance, and adaptation.
Today
- Compute = clock cycles and transistors
- Energy cost grows faster than efficiency
- AI workloads scale linearly with silicon
Emerging
- Analog & neuromorphic systems rediscover efficiency
- Quantum offers non-linear scaling but limited practicality
- Hybrid bio-materials show stable room-temp coherence
Next
- Computation becomes material — not simulated in silicon
- Energy landscapes replace instruction sets
- Self-assembled lattices become the new hardware fabric
In simple terms: ProQ doesn’t run algorithms; it is one. Each lattice behaves as a tiny physical solver, exploring possibilities through quantum-analog dynamics. As compute transitions from bits to behaviors, ProQ defines the material foundation of this new paradigm.
Partner with ProQ
We collaborate with labs and platform partners exploring inference efficiency, quantum-aware sensing, and hybrid compute modules.
Press & briefings
For media or analyst briefings on the age of inference and quantum-analog substrates, get in touch.