Applicants

My PhD research examines sediment connectivity in the Ondava River, focusing on sediment transfer, storage, and channel adjustment. Using geomorphology, GIS, and spatial analysis, I strive to support scientific river management, restoration planning, and hazard assessment in dynamic fluvial systems.

My PhD research investigates sediment connectivity in the Ondava River in eastern Slovakia. The goal is to explain how sediment is entrained, transferred, intermittently stored, and remobilized within a gravel-bed fluvial system under varying hydrogeomorphic and anthropogenic controls.

A fundamental control on channel evolution, geomorphic sensitivity, ecological condition, and river management response. By integrating field geomorphology, GIS, and quantitative analysis.

Artificial intelligence is the breakthrough in the field of geoinformatics whose automatisation quality is really rigorous and helps to improve the data analysis.

[No response provided]

I aim to solve the problem of safe, sensitive, real‑time detection of environmentally and industrially for hazardous gases where radioactive sources are undesirable or restricted.

By developing a corona‑discharge IMS that delivers high, stable ionization without radioactivity for monitoring.

My project develops a non‑radioactive ion mobility spectrometer that uses corona discharge to generate ions for real‑time gas analysis. It targets sensitive, on‑site monitoring of environmentally and industrially relevant hazardous gases where traditional radioactive sources are impractical or restr

My idea replaces radioactive IMS ion sources with a compact corona‑discharge source that delivers higher ion yields and tunable ion chemistry. This boosts sensitivity and selectivity while removing licensing, safety, and disposal burdens.

Industrial workers, safety officers, and environmental agencies gain safer, real‑time monitoring of hazardous gases without radioactive sources, while instrument makers get a simpler, regulation‑friendly IMS platform for broader field deployment.

Green hydrogen is expensive due to an “efficiency tax”. Standard electrodes use polymer glues (binders) that block active sites and degrade in harsh alkaline conditions. This causes massive energy losses and high maintenance, preventing clean fuel from scaling globally.

By developing ATLAS (Atomic-Tuned Lattice for Alkaline Splitting), I leverage High Entropy Oxides to create a binder-free, self-supporting electrode. This atomic-scale engineering ensures maximum energy flow and structural integrity without the need for fragile chemical glues.

We propose ATLAS, a blueprint for the next generation of industrial electrodes. By transitioning from “painted-on” powders to integrated, conductive HEO structures, ATLAS provides a robust and efficient platform for commercial-scale hydrogen production.

Unlike standard designs, ATLAS turns the catalyst into the electrode itself. By harnessing the “atomic chaos” of five elements, this project explores a unique synergy that allows Earth-abundant oxides to outperform noble metals like Platinum in both durability and cost.

ATLAS makes green hydrogen a profitable reality. By aiming for 20% longer lifespans and 15% lower operating costs, it removes the price barrier for carbon-neutral steel, shipping, and aviation, transforming clean energy from a future goal into an immediate industrial solution.

We owe permanent implants a lot—they have transformed bone repair and remain essential in many clinical situations. But when an implant’s role is only to support healing, why should it remain in the body long after its job is done?

My research explores a new generation of biodegradable biomaterials that provide temporary support for bone healing and gradually disappear once their job is done. Instead of permanently replacing tissue, these materials work with the body, supporting its natural ability to regenerate.

Using stem cells, I compare biodegradable biomaterials to understand how their intrinsic properties influence bone regeneration. By identifying materials that actively support the body’s natural healing processes, my research aims to advance the next generation of regenerative temporary implants.

For decades, we have celebrated implants that last forever. But perhaps the future belongs to implants designed to safely disappear after healing is complete. My research challenges this idea by developing biomaterials that support regeneration before naturally degrading.

The project could help redefine the role of implants in regenerative medicine—from permanent implants to temporary partners that support healing and leave only regenerated tissue behind.

Creating large-scale physical structures from a digital idea requires navigating a fragmented chain of incompatible tools — CAD, slicers, robot programming environments. This friction kills creative momentum and locks innovation behind technical gatekeepers.

Grasshopper — a visual programming sandbox — becomes the single environment where design, simulation, material logic, and six-axis robot toolpath generation all coexist and interconnect. One idea flows uninterrupted from concept to a physical large-format 3D printed object.

The pipeline connects computational geometry, process parameters, and robot controlling. Anyone fluent in algorithmic thinking is enabled to fabricate physical objects.

Grasshopper was designed as a sandbox where anything can connect to anything. We exploited exactly that — linking design algorithms, material science constraints, and FANUC robot kinematics to fabricate physical objects.

When the barrier between idea and production disappears, a new generation of designers, engineers, and researchers will be able to create prototypes and produce things that previously required entire teams of specialists, thereby accelerating innovation in architecture, industry, and science.

International law permits lethal force only where human judgment, legal justification and accountability remain possible. Autonomous weapons challenge this premise by shifting target selection to opaque systems, creating risks of unlawful death without an identifiable legal answer.

A legal framework is proposed to determine when lethal functions may be delegated to autonomous weapons. It translates the right to life, IHL and state responsibility into tests of human control, command traceability, auditability, targeting review and enforceable liability.

This project asks whether a state may lawfully delegate lethal decision-making to autonomous weapons without undermining the protection of life. It proposes a legal framework defining who must control, justify and answer for harm caused by such systems.

The innovation lies in reframing autonomous weapons from a question of technical accuracy to one of legal attribution, review and accountability. It improves current approaches by turning the right to life, IHL and state responsibility into concrete tests for control, traceability and liability.

Lawmakers, courts, international bodies and military decision-makers benefit from clear legal criteria for regulating autonomous weapons. Civilians and victims benefit through stronger accountability, evidentiary duties and access to meaningful review after harm.

Human beings, in this case judges, can be easily influenced, regardless of how strongly they claim otherwise. Merely knowing the identities of the parties to a legal dispute may lead a judge to make a decision that is not entirely objective, particularly if one of the parties is a public figure.

Anonymizing the parties during the initial stages of the proceedings can enhance judicial impartiality and thereby contribute to reaching an objective decision.

Anonymization of the parties to a legal dispute during the initial stages of court proceedings. Enhanced judicial impartiality.

Court proceedings have been based on the same principles and procedures for centuries, while complaints about judges regarding their impartiality are extremely numerous. One possible reform is to anonymize the parties to a dispute at least in the initial stages of the proceedings.

It may primarily benefit the parties to the dispute, who would be to some extent better protected against potential judicial bias. This approach could also increase the likelihood of reaching a fairer decision and help avoid unnecessary procedural steps, such as filing complaints.

Scaling continuous-time Al (Neural ODEs) to human-scale (8B nodes) leads to crashes. High-precision solvers hit a massive memory wall, while cheap, low-precision hardware introduces rounding errors that trigger catastrophic numerical explosions in standard Euler schemes.

Introducing a novel State-Dependent Stochastic-Informed integrator, which mathematically splits neural dynamics, directly overwriting the massive state vector without new memory allocation need, crushes memory overhead down to zero, unlocking human-scale models on standard consumer hardware.

We developed a hardware-fused mathematical integrator that stabilizes large-scale Neural ODEs against numerical divergence in stiff regimes. This framework eliminates memory overhead, allowing 8-billion-parameter continuous Al to run stably on a standard consumer hardware.

Current Al relies on expensive supercomputers or memory-heavy solvers. Our breakthrough lies in an adaptive, state-dependent contraction mechanism that controls the dissipative contribution during spikes, thereby infinitesimal point-wise accuracy is traded for the total global structural stability.

Developers and independent labs can benefit by bypassing million-dollar clusters, deploying human-scale, continuous-time Al directly on standard consumer hardware. This eliminates cloud costs and latency, enabling private, always-on personal agents or real-time edge diagnostics.

Cadmium contamination in croplands allows toxic metal to enter food crops, especially edible seeds, and microplastics may worsen plant stress and uptake. This threatens food safety on polluted farmland, where full soil remediation is slow and costly.

Spray lentil leaves with green-synthesized nanoparticles made from mushroom-farming waste to activate the plant’s defense and reduce cadmium movement into edible seeds. This crop-protection approach is low-cost, uses local waste, and is designed for polluted farmland where remediation is difficult.

I develop foliar nanoparticles from mushroom-farming waste to reduce cadmium transfer into lentil seeds under combined cadmium and microplastic stress. The project tests a crop-protection strategy that can improve food safety on contaminated croplands without waiting for full soil remediation.

The breakthrough is shifting from soil cleanup to crop protection. Instead of removing all contamination from land, foliar nanoparticles from mushroom waste help the plant block cadmium uptake and protect edible seeds, offering a low-cost, locally adaptable alternative.

Farmers and consumers on contaminated land benefit most. The project aims to lower cadmium in edible lentil seeds, improve food safety, and offer a low-cost crop-protection strategy using farm waste, especially where full soil remediation is too expensive.

Much scientific effort is devoted to the use of nanomaterials in biomedicine, yet current methods have not been precise at the nanoscale. We still lack a documented, mechanistic account of what actually happens to a nanoparticle after its entry into a cell.

The chemical signals of MoOx nanoparticles change in response to their environment. This responsivity can be used to investigate distinct phases of nanoparticle transformations in cell. Novel near-field imaging enables tens-of-nanometers resolution, providing a direct view into MoOx cell processing.

MoOx nanoparticles are localized at the nanoscale by scanning scattering near-field optical microscopy at a characteristic infrared vibration. Cell-induced changes in FTIR spectra of the MoOx are tracked by nano-FTIR at defined time points after the incubation, elucidating their metabolomics.

The project combines state-of-the-art chemical tracking in cells on the nanoscale with environmentally responsive MoOx nanoparticles. It represents the first use of nano-resolved chemical tracking to study cell-induced transformations of nanoparticles in detail.

Mapping cell-induced transformations of nanoparticles sets the stage for smart design of next-generation nanomedicines applicable to targeted drug delivery, immunomodulation or photothermal treatment. Simultaneously, it unlocks the full potential of MoOx‑based precision photothermal therapy.