Blab Business
MagiQware Secures 575,000 Pre-Seed Funding to Reduce Quantum Computing Overhead
MagiQware, a spin‑off from Delft University of Technology, closed a pre‑seed funding round of €575,000 ($658 k) on July 6 2026. The capital, led by early‑stage investor LUMO Labs and supplemented by Graduate Ventures and Delft Enterprises B.V., will be used to accelerate product development, validate automated quantum compilers, and expand the company’s engineering team.
The round began on May 28 2026 with a €500 k close under the TTT.AI programme. A subsequent co‑investment by Graduate Ventures and Delft Enterprises B.V. added €75 k, bringing the total to €575 k. The funding is earmarked for the development of MagiQware’s reinforcement‑learning‑driven software stack, which targets the most costly component of fault‑tolerant quantum computing (FTQC): magic state factories.
FTQC requires quantum error correction to suppress decoherence. While logical qubits can be protected by surface codes and other stabilizer codes, the implementation of non‑Clifford logical gates—necessary for universal computation—depends on magic state distillation (MSD). MSD circuits consume a large fraction of a quantum processor’s physical qubits and gate operations; estimates place the overhead at up to 90 % of the total qubit and circuit footprint.
MagiQware’s solution is to embed reinforcement‑learning agents within the quantum compiler. The agents automatically design and optimise distillation circuit architectures, discovering configurations that reduce circuit length and resource usage. Early benchmarks from the company’s internal tests show a reduction of up to 40 % in the number of gates required for target factories, without any changes to the underlying hardware.
The company’s technical leadership includes CEO Arash Ahmadi, PhD, CTO Shakeeb Majid, Head of Device Sahar Hejazi, PhD, and Head of Theory Ali Moghaddam, PhD. Together, they oversee a team that combines expertise in quantum algorithms, machine‑learning optimisation, and hardware‑aware compiler design.
According to reports from LUMO Labs, the investment will support the validation of the automated compiler pipeline and the scaling of the reinforcement‑learning models to larger, more complex distillation tasks. Graduate Ventures noted that the capital will also fund the expansion of the engineering workforce, enabling MagiQware to accelerate its roadmap toward commercial deployment.
The broader context of the funding is the ongoing race to build practical, large‑scale quantum computers. While current noisy intermediate‑scale quantum (NISQ) devices can perform limited tasks, they lack the error‑correction capabilities required for general‑purpose applications. The transition to FTQC hinges on reducing the physical resource overhead of error‑correction, of which MSD is a critical bottleneck.
MagiQware’s approach aligns with industry efforts to make quantum hardware more efficient. By reducing the qubit and gate overhead associated with magic state factories, the company aims to lower the cost and complexity of future quantum processors. The company’s technology is positioned to complement hardware advances from firms such as PsiQuantum and other silicon‑photonic or superconducting platforms.
The €575 k pre‑seed round places MagiQware among a growing cohort of quantum‑software startups that are leveraging artificial intelligence to tackle hardware‑level challenges. The company has made its milestones and commercial roadmap publicly available through institutional disclosure documents, and it has announced plans to collaborate with academic and industrial partners to validate its solutions on emerging quantum hardware.
At present, MagiQware remains in the pre‑seed stage, with no announced revenue streams or customer contracts. The company’s next milestones include the release of a beta version of its compiler suite, the completion of a proof‑of‑concept demonstration on a commercial quantum platform, and the scaling of its reinforcement‑learning models to support larger distillation factories.
The funding round is expected to enable MagiQware to move from prototype to a more mature product, potentially attracting additional investment from venture capital or strategic partners in the quantum ecosystem. The company has not yet disclosed a timeline for a subsequent funding round or for commercial product launch.
In summary, MagiQware’s €575 k pre‑seed investment provides the capital needed to advance a reinforcement‑learning‑based compiler that targets a key bottleneck in fault‑tolerant quantum computing. The company’s progress will be closely watched by stakeholders in the quantum industry, as reductions in magic state distillation overhead could accelerate the deployment of practical, large‑scale quantum processors.
The round began on May 28 2026 with a €500 k close under the TTT.AI programme. A subsequent co‑investment by Graduate Ventures and Delft Enterprises B.V. added €75 k, bringing the total to €575 k. The funding is earmarked for the development of MagiQware’s reinforcement‑learning‑driven software stack, which targets the most costly component of fault‑tolerant quantum computing (FTQC): magic state factories.
FTQC requires quantum error correction to suppress decoherence. While logical qubits can be protected by surface codes and other stabilizer codes, the implementation of non‑Clifford logical gates—necessary for universal computation—depends on magic state distillation (MSD). MSD circuits consume a large fraction of a quantum processor’s physical qubits and gate operations; estimates place the overhead at up to 90 % of the total qubit and circuit footprint.
MagiQware’s solution is to embed reinforcement‑learning agents within the quantum compiler. The agents automatically design and optimise distillation circuit architectures, discovering configurations that reduce circuit length and resource usage. Early benchmarks from the company’s internal tests show a reduction of up to 40 % in the number of gates required for target factories, without any changes to the underlying hardware.
The company’s technical leadership includes CEO Arash Ahmadi, PhD, CTO Shakeeb Majid, Head of Device Sahar Hejazi, PhD, and Head of Theory Ali Moghaddam, PhD. Together, they oversee a team that combines expertise in quantum algorithms, machine‑learning optimisation, and hardware‑aware compiler design.
According to reports from LUMO Labs, the investment will support the validation of the automated compiler pipeline and the scaling of the reinforcement‑learning models to larger, more complex distillation tasks. Graduate Ventures noted that the capital will also fund the expansion of the engineering workforce, enabling MagiQware to accelerate its roadmap toward commercial deployment.
The broader context of the funding is the ongoing race to build practical, large‑scale quantum computers. While current noisy intermediate‑scale quantum (NISQ) devices can perform limited tasks, they lack the error‑correction capabilities required for general‑purpose applications. The transition to FTQC hinges on reducing the physical resource overhead of error‑correction, of which MSD is a critical bottleneck.
MagiQware’s approach aligns with industry efforts to make quantum hardware more efficient. By reducing the qubit and gate overhead associated with magic state factories, the company aims to lower the cost and complexity of future quantum processors. The company’s technology is positioned to complement hardware advances from firms such as PsiQuantum and other silicon‑photonic or superconducting platforms.
The €575 k pre‑seed round places MagiQware among a growing cohort of quantum‑software startups that are leveraging artificial intelligence to tackle hardware‑level challenges. The company has made its milestones and commercial roadmap publicly available through institutional disclosure documents, and it has announced plans to collaborate with academic and industrial partners to validate its solutions on emerging quantum hardware.
At present, MagiQware remains in the pre‑seed stage, with no announced revenue streams or customer contracts. The company’s next milestones include the release of a beta version of its compiler suite, the completion of a proof‑of‑concept demonstration on a commercial quantum platform, and the scaling of its reinforcement‑learning models to support larger distillation factories.
The funding round is expected to enable MagiQware to move from prototype to a more mature product, potentially attracting additional investment from venture capital or strategic partners in the quantum ecosystem. The company has not yet disclosed a timeline for a subsequent funding round or for commercial product launch.
In summary, MagiQware’s €575 k pre‑seed investment provides the capital needed to advance a reinforcement‑learning‑based compiler that targets a key bottleneck in fault‑tolerant quantum computing. The company’s progress will be closely watched by stakeholders in the quantum industry, as reductions in magic state distillation overhead could accelerate the deployment of practical, large‑scale quantum processors.
