
About me
I'm Roman, a Lead AI/ML Engineer with 15+ years of software development. I take AI products from idea to production: LLM agents, RAG, text-to-SQL and MCP servers, with the backend, data, integrations, cloud and security behind them. I also do a lot of R&D and project estimation, work with clients as a Forward-Deployed AI Engineer, set engineering standards and mentor developers. Before that I built web, backend and mobile applications for banks, government and business.
Now
Agents and multi-agent systems for concrete business tasks:
- agents embedded in enterprise software and wired into its live backends
- voice and multimodal agents
- natural-language analytics over company databases (text-to-SQL)
- RAG over internal knowledge and documents
- MCP servers and orchestration
- LLM deployment and scaling, all the way down to local devices
I also build projects designed from the start for agent-driven development, where the whole cycle — analysis, development, QA, review, issue tracking — runs through role-configured agents; still rare in practice. I own the architectural baseline for the team's projects and run discovery and estimation on new ones.
Before
PHP since 2009, Java backend and web since 2010, Android since 2018. Microservice architectures in Java, Kotlin and Python — mission-critical systems serving millions of users, legacy landscapes, regulatory constraints.
How I work
Equally effective on my own, in a team, or as lead developer. I take projects from architecture to release in full; in a team I own the architectural frame and the technical standards. Demanding about code quality, but never at the expense of deadlines. I switch contexts quickly and hold several systems in detail at once. Moved into AI early, at GPT-3.5, when the field was still forming.
Links
Skills
- AI & LLM
- AI agents
- multi-agent systems
- RAG
- text-to-SQL
- MCP
- LLM evaluation
- voice and multimodal AI
- Frameworks
- LangGraph
- LangChain
- DSPy
- FastAPI
- Spring
- Models & ML
- OpenAI
- Claude
- Gemini
- open-source and on-device LLMs
- PyTorch
- fine-tuning (LoRA)
- Programming
- Python
- TypeScript
- Kotlin
- Java
- SQL
- Backend
- REST and gRPC APIs
- WebSocket
- microservices
- message brokers (Kafka, RabbitMQ)
- clean architecture
- Data
- PostgreSQL
- SQL Server
- MySQL
- MongoDB
- Redis
- vector databases
- Cloud & DevOps
- AWS
- Azure
- Google Cloud
- Docker
- Kubernetes
- Terraform
- CI/CD
- monitoring
- Security
- authentication and authorization (OAuth)
- encryption
- secrets management
- PII protection
- AI tooling
- Claude Code
- Cursor