Observability is not just monitoring
Classic monitoring answers the question “is the server running.” But a digital service is a chain: user action, application, database, infrastructure. Every link can be formally operational, yet the service does not work.
A practical example: a customer cannot pay for an order even though all servers are fine. Observability is needed precisely for such cases: the platform links technical metrics with specific operations and shows where the problem occurred, which services it affected, and what to check to restore operation.
Glossary
Metrics — numerical indicators of system performance: load, response time, error count.
Event logs — records of what happened in the system.
Traces — data about a request's path through system components.
Resource-service model — a map of relationships between applications and the infrastructure on which they run.
How AI helps investigate incidents
BOSS correlates metrics, logs, and traces and automatically builds a resource-service model. On this basis, AI helps find the probable cause of the incident, reconstruct the failure chain, and suggests recommendations for remediation.
• Anomaly detection. Machine learning algorithms account for seasonality — a typical evening load spike is not considered an issue.
• Less noise. Related alerts are grouped into a single incident rather than arriving as dozens of separate notifications.
• Chat instead of queries. A specialist can ask in plain language which services are affected by an outage or why an application's response time has increased.
• Custom language model. For AI features, a model with an OpenAI-compatible interface can be connected, including one deployed on the customer’s premises — data does not have to leave the perimeter.
According to the developers' estimate, such an approach can reduce investigation time from days to minutes.
A business perspective
The second half of the product is addressed to executives. The platform assesses the state of IT at the level of business units: technical metrics are linked to business systems, processes, and operations, making it clear which of them were affected by a failure.
Analysis of user sessions and journey maps shows where people experience difficulties and which actions they cannot complete. In addition, the platform monitors service quality targets and the error budget.
How well does this pay off? A benchmark is provided by the international New Relic 2024 Observability Forecast study: companies with full-stack observability experience 79% less total annual downtime on average, and an hour of downtime costs 48% less.
What BOSS is built on
The technological foundation is Proto Observability Platform. In 2025, analysts at Technologies of Trust named Proto one of the key developers in the field of observability, and in September 2026, Basis acquired 70% of the company.
According to the developers' estimate, 95% of the functionality of key foreign solutions in this class — Dynatrace, Datadog, and New Relic — is already implemented in the first version.
BOSS can be implemented as a standalone product into existing infrastructure. For those who already use Basis solutions, native integration with the vendor’s ecosystem is available.
“We created BOSS together with the Basis team, combining our operational data analysis and incident investigation technologies with the vendor’s infrastructure expertise. Already in the first version, the platform automatically builds a model of relationships between applications and infrastructure, correlates metrics, and uses AI for failure diagnostics,” — Denis Bezkorovayny, CEO of Proto
Who needs this
First and foremost — organizations for which controlling the entire operational chain of a digital service is critical. The company specifically highlights the financial and public sectors.
What is happening in the market
According to data from the iKS-Consulting study “Dynamic IT Infrastructure Management Software Market (DIM Software) 2026”, infrastructure is becoming more complex, and a unified management stack is forming around it: virtualization, containers, and cloud platforms along with end-to-end observability and operations automation. An additional boost is provided by AI: language models and agents are themselves becoming observability targets and increasing the volume of telemetry.
“As IT infrastructure becomes more complex, the need for end-to-end diagnostics using AI will grow. We view observability as one of the avenues for further growth of Basis. Further work on the product will focus on refining AI diagnostics and deeper integration with our solutions,” — David Martirosov, CEO of Basis
Important considerations
“95% functionality” and “from days to minutes” are developer estimates, not the results of an independent comparison.
The New Relic figures are average values from an international sample; the result of a specific company depends on the maturity of its operational processes.
About the company
Basis is a Russian software developer for dynamic IT infrastructure management, virtual desktops, and cloud services. The company was founded in 2021 by combining the assets of Tionix, Digital Energy, and Skala Software, and is controlled by the RTK-TsOD group. Since December 2025, its shares have been traded on the Moscow Exchange under the ticker BAZA. The products are included in the Unified Register of Russian Software and certified by the FSTEC of Russia.
Learn more about the products on the company's website: basis.ru
erid: 2VtzqumWYyL Advertisement. OOO “Basis” INN 7731316059