- Ukraine’s Central Research Institute has unveiled Marichka, a closed neural network aimed at reducing brigade-level planning decisions from six to 12 hours to minutes.
- The system calculates force ratios, estimates losses, and models courses of action on an interactive map, but so far it has only been tested with test data, and integration of classified documents is still pending.
- Marichka adds another layer to Ukraine’s technology stack for battlefield planning: the country already has Delta for situational awareness and works with Palantir on multiple levels.
The Central Research Institute of the Armed Forces of Ukraine has introduced Marichka, a closed neural network designed to compress the operational planning cycle at the brigade level and above from hours to minutes.
The system, still in testing, is designed to collect and verify battlefield information, then calculate force ratios and estimate necessary resources and likely losses, presenting commanders with options modeled on an interactive map. Currently, military standards allow a regiment or brigade commander to need six to 12 hours and approximately 20 documents to get to the same point.
It can also alert command positions to changes in the operational situation as new information arrives. Col. Yurii Kliat, who heads the institute, was explicit about where authority ends, saying at the briefing that the final decision rests with the commander.
Latest videos ofTechnologyRadar
A battlefield cape replacement for Palantir
Palantir has been integrated into Ukrainian operations since shortly after Russia’s full-scale invasion in 2022, and remains a mainstay in that theater, with a reach that dwarfs what Marichka is attempting.
Marichka remains an important step. If it works as described, it would give Ukraine sovereign control over part of the reasoning layer currently in the hands of a foreign supplier, even as the Palantir pipeline remains the most established presence in the country’s software-driven war effort.
According to Colonel Yurii Kliat, head of the Central Research Institute of the Armed Forces of Ukraine, Marichka centrally collects information about the battlefield, automatically calculates the balance of forces and combat potential, and estimates the necessary resources and possible losses.
The intent is to allow brigade and regimental commanders to make an informed decision in minutes, compared to a doctrinal standard that currently allows six to 12 hours and approximately 20 documents.
Ukraine already has the system that comes closest to describing a local Palantir: Delta, which has provided situational awareness to all defense forces since 2022. Marichka is a proposed new layer on top of a stack that already includes Delta, Palantir’s platforms for fusion and analysis, and, starting this year, a dedicated Defense Artificial Intelligence Center. It hasn’t been revealed whether it will be integrated with Delta or run separately, and that is possibly the biggest open question about the program.
Brigadier General Andrii Lebedenko said at the Drone Autonomy 2026 conference in May that the Armed Forces had already begun integrating AI agents into command and control systems, placing Marichka within an existing effort rather than presenting it as a way out.
This is not Ukraine’s first attempt to create an operational-level command system for the headquarters. Dzvin was commissioned to a contractor in 2016 and frozen in 2021 after prototypes faced disputes over costs, timelines and product ownership.
Delta, developed in parallel and not as a replacement, was successful in part because it emerged from a volunteer reconnaissance unit that solved an immediate problem on the battlefield, rather than from an acquisition process.
Marichka comes from a research institute, which also has a different pedigree, and the most difficult test will not be the neural network but the bureaucracy around the accreditation of classified data. That obstacle may be reinforced by the argument in favor of the current president: Palantir has a proven track record in theaters of war, and its early work in Ukraine, much of it pro-bono, ensured a first-mover advantage that a domestic prototype will have to overcome.
Via United24Media
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds.
