Making complex systems
intelligible
We design and deploy production-grade generative AI: knowledge systems, AI agents, private LLMs. Our deep specialization is IT operations, where complexity and failure are hardest to reason about.
Who we are
Trilagi is a deep-tech startup founded by three co-founders with extensive experience in AI research, distributed systems, and large-scale R&D. We build intelligent systems that reason about complex technical environments instead of just automating them.
What we build
Production-grade GenAI implementations in the cloud, on-premise, or fully managed by us, with your data always under your control. Assembled from the same building blocks we rely on in demanding IT operations environments.
- AI Knowledge SystemsFind anything. Instantly. Across all your internal knowledge.
- AI Agents & AutomationReplace repetitive workflows with AI agents that think, decide, and act.
- AI Communication & AssistantsAn AI assistant your team will actually use, right where they already work.
- Data Privacy & Secure AIUse AI on sensitive data without the compliance risk.
- GenAI Infrastructure & DeploymentOwn your AI stack. Run models on your terms.
- All services
AI for IT operations
GenAI is our craft; IT operations is where we go deepest. Cloud and hybrid platforms have become too complex to reason about under failure. Signals are fragmented, context is lost, and diagnosis remains largely manual: slow, error-prone, and expensive.
Working alongside operations engineers on live infrastructure, we build AI that forms an internal model of the environment it works in. That lets it reason about behavior, change, and failure the way an experienced engineer does.
Semantic understanding of heterogeneous technical data
Logs, metrics, tickets, runbooks and configuration read as one connected picture instead of separate silos.
Reasoning over system behavior, not just signals
Connecting what changed, what broke and why, so diagnosis starts from a hypothesis, not a search.
Hierarchical AI agents collaborating on complex tasks
Specialized agents gather evidence, test hypotheses and hand engineers a structured, verifiable analysis.
Our product
RCA Assistant
A root cause analysis assistant for ITOps, DevOps and AIOps teams, built on these capabilities.
Research foundations
Our work bridges computational linguistics, semantic processing of heterogeneous technical data, and distributed hierarchical multi-agent architectures. We work on how AI systems recognize cause and effect and diagnose failures. What proves itself in research goes into the systems we deliver to clients.
This research is part of an industrial R&D project co-funded by the European Union. About the project
Meet the Team
Marcin Krysiński
CVO, Lead AI/LLM Engineer, Co-Founder
Seasoned AI/ML/DL/LLM expert with a proven track record in delivering innovative AI solutions and leading large-scale R&D initiatives.
Damian Szewczyk
CEO, Co-Founder
Leads the company's strategy and business development, combining experience in IT service management with a practical approach to technology in business.
Mariusz Nowosielski
CTO, Lead Cloud Engineer, Co-Founder
Cloud and infrastructure expert specializing in private, public, and hybrid cloud environments and large-scale system implementations.
Let's talk
Have a project or research idea? Let's book a 30-minute call and see what we can do together.
Meetings are booked through Microsoft Bookings. See how we handle your data in our privacy policy.