Roman Kurnovskii

Mr. Roman Kurnovskii

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 September 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
RKRoman Kurnovskii
Roman Kurnovskii
Concentrated liquidity market makers (CLMMs), such as Meteora's Dynamic Liquidity Market Maker (DLMM) on Solana, require liquidity providers to continuously select pools, size discrete bin ranges, and time entries and exits. This constitutes a sequential decision problem under partial observability that static, parametric rebalancing vaults (e.g., Kamino Finance, 2024; Gamma Strategies) handle poorly in volatile, high-turnover markets. Conversely, recent large language model (LLM) trading agents (e.g., FinMem (Yu et al., 2023), FinAgent (Zhang et al., 2024)) focus predominantly on isolated single-agent spot or perpetual trading, lacking domain-specific bin mechanics and cross-instance knowledge sharing. We present Etemaro, an autonomous architecture that utilizes an LLM inside a ReAct reasoning loop to screen candidate pools, deploy and manage DLMM positions, and execute exits based on live profit-and-loss, yield, and bin drift data. Etemaro introduces three primary contributions beyond standard trading-agent frameworks: (1) a Darwinian signal-weighting mechanism that reweights heuristic screening signals online using a quartile-lift statistic computed from realized win/loss outcomes; (2) role-scoped and intent-filtered tool access, strictly constraining which on-chain actions the LLM may execute across screening, management, and chat contexts; and (3) HiveMind, a lightweight federated experiential-learning protocol where independent agent instances exchange natural-language lessons and scored strategy presets without sharing raw transaction logs, private keys, or model weights. We evaluate Etemaro through a continuous 114-day production deployment on Solana Mainnet (April 1 to August 24, 2026). Across this period, the agent executed 4,583 on-chain deployments committing 9,715.52 SOL in capital (~2.12 SOL average per position) and generated 50,329 on-chain lifecycle events, achieving a 68.35% realized win rate (743 wins out of 1,087 closed positions) with a median hold time of 36.5 minutes. We candidly document operational failure modes, including a 37.25% state divergence between local logs and on-chain reality caused by RPC dropouts and 5,521 concurrent race-condition transactions. We position this work as a systems and architecture contribution, establishing an empirical template for safety-constrained, self-improving autonomous agents operating in live financial environments. Overall, our findings demonstrate that combining bounded reasoning loops with lightweight experiential federation enables autonomous agents to operate effectively and self-adaptively in high-velocity decentralized financial markets.

16 September 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
OUOscar Underwood
Oscar Underwood
RKRoman Kurnovskii
Roman Kurnovskii
Autonomous software agents are increasingly transacting over the HTTP 402 (x402) protocol, yet the empirical economics of machine-to-machine commerce remain largely uncharacterized. The prevailing industry narrative assumes a migration toward multi-dollar, compound transactions. To test this, we analyzed a census-scale dataset of 143.8 million x402 settlements—totaling $42.8M across Base, Solana, and Polygon (May–Dec 2025)—and found the exact opposite. Our analysis reveals three structural realities of the machine economy: The Sub-Dollar Reality: Machine commerce is overwhelmingly micropayment-driven. The volume-weighted average payment is just $0.27, and 28.5% of all transactions are sub-cent. Payments ≥$1 carry 72.5% of the total value but represent less than 8% of network traffic. Extreme Centralization: The top 1% of merchants capture 93.0% of all revenue. This bottleneck is even more pronounced at the settlement infrastructure layer, where a single facilitator controls ~62% of the total settled volume. Wild Price Dispersion & Network Dominance: Endpoint list prices are highly erratic (median $0.01, CV 12.91), decisively rejecting the hypothesis of a tight, efficient algorithmic clearing corridor. Furthermore, settlement is highly skewed toward Base (82.5% of volume), with Solana holding a persistent minority (17.4%). These findings dismantle the myth of complex machine commerce, reframing the x402 ecosystem as a high-volume, sub-dollar API economy where the true concentration risk lies in settlement infrastructure, rather than among merchants.

1 November 2026 in Preprints.ru
Verified
Preprint
Open access
НКНикита Курновский
Никита Курновский
RKRoman Kurnovskii
Roman Kurnovskii
В сельском хозяйстве России средняя заработная плата выросла в 2,25 раза за 2018–2024 гг. — с 25 808,4 до 58 177,4 руб., но осталась на уровне 58,8 % средней по экономике (организации без субъектов малого предпринимательства; 51,7 % в 2018 г.), а абсолютный разрыв вырос с 24 086,2 до 40 803,2 руб.; в условиях дефицита кадров эти различия значимы для политики удержания. Цель работы — описать дифференциацию, внутригодовую сезонность и региональные различия оплаты труда в отрасли по открытым данным Росстата. Аналитические окна различаются: разрыв — 2018–2024 гг., реальная заработная плата — 2017–2025 гг., сезонность — 2023–2024 гг., региональный разрез — 2024 г., поддержка и инвестиции — 2017–2023 гг.; период 2017–2025 гг. в названии отражает доступность данных. Относительный разрыв сократился на 7,1 п.п.; реальная заработная плата в ценах 2024 г. выросла на 52,2 % за 2017–2024 гг.; внутригодовая амплитуда в 2024 г. составила 41,5 % между январём и декабрём и 44,3 % между максимумом и минимумом, после удаления тренда — 14,9 %; региональная медиана — 58 518,6 руб. при отношении максимума к минимуму 6,71. Вывод: различия в оплате устойчивы во времени и пространстве и задают контекст для политики удержания персонала.

1 January 2024 in International Journal of Applied and Fundamental Research (Международный журнал прикладных и фундаментальных исследований)
Verified
Journal Article
Open access
RKRoman Kurnovskii
Roman Kurnovskii

23 October 2023 in Экономика и предпринимательство
Verified
Journal Article
Closed access
RKRoman Kurnovskii
Roman Kurnovskii
ЕВЕ.а. Великородная
Е.а. Великородная
Цепочка создания стоимости (иными словами, отраслевая цепочка создания стоимости) состоит из экономических субъектов, предоставляющих необходимую компании ресурсную составляющую, и их отдельно взятых цепочек создания стоимости. Появившись в средине в 1985 года в трудах Майкла Портера в качестве модели компании, представляющей ее в формате основных и вспомогательных процессов, на сегодняшний день цепочка создания стоимости получила широкое признание у специалистов и консультантов, работающих в сфере стратегического управления. Целью статьи выступает рассмотрение содержания, ключевых элементов и возможностей, позволяющих использовать цепочку создания стоимости. The value chain (in other words, the industry value chain) consists of economic entities that provide the necessary resource component to the company, and their individual value chains. Having appeared in the middle in 1985 in the writings of Michael Porter as a model of a company, representing it in the format of basic and cognitive processes, today the value chain has been widely recognized by specialists and consultants working in the field of strategic management. The purpose of the article is to consider the content, key elements and opportunities that allow using the value chain.

10 June 2026
Verified
Journal Article
Closed access
RKRoman Kurnovskii
Roman Kurnovskii
EVEkaterina Velikorodnaia
Ekaterina Velikorodnaia