Founder & CTO building private, agentic and sovereign AI systems for enterprise and edge environments.
I am a Senior Software Engineer and AI Platform Architect with 10+ years of experience designing distributed systems, cloud-native platforms and production-grade AI applications.
I am currently building SayMyData, a private agentic AI platform that helps organizations transform trusted data, documents and operational systems into actionable knowledge through specialized agents, local AI models and governed tool execution.
SayMyData is designed for enterprise, industrial and regulated environments where AI must remain private, auditable and operationally useful.
The platform combines:
- Maestro orchestration
- Collaborative agents
- Specialist agents
- Knowledge Hub
- Structured business concepts
- Runtime memory
- Tool calling
- Policy enforcement
- Local LLM / SLM inference
- Fine-tuning and model deployment
- Edge deployment on NVIDIA DGX Spark / GB10-class hardware
- Cloud fallback when needed
The goal is to move beyond a single chatbot and build an AI runtime where each mission is executed by the right agent, the right model and the right tool.
I am currently working on:
- Multi-agent orchestration
- Agentic runtime design
- Local LLM and SLM deployment
- NVIDIA DGX Spark / Blackwell inference
- vLLM, NVIDIA NIM and local serving pipelines
- BF16, FP8 and NVFP4 model deployment
- LoRA / PEFT fine-tuning
- Specialist agents for energy, maintenance, knowledge and operations
- Knowledge Hub and structured concepts
- BM25 / sparse retrieval without mandatory dense embeddings
- Tool calling and structured outputs
- Runtime memory and execution traces
- Edge and sovereign AI infrastructure
- Go
- TypeScript
- Node.js
- Python
- REST APIs
- Microservices
- Event-driven architectures
- NATS JetStream
- Redis
- PostgreSQL
- MinIO / object storage
- Kong Gateway
- Large Language Models
- Small Language Models
- Agentic AI
- Retrieval-Augmented Generation
- Sparse retrieval
- BM25 full-text search
- Knowledge systems
- Tool calling
- Structured outputs
- Prompt engineering
- LoRA / PEFT
- Model evaluation
- NVIDIA NeMo
- vLLM
- NVIDIA NIM
- TensorRT-LLM
- Docker
- Docker Compose
- Kubernetes
- AWS
- EKS
- CI/CD
- Observability
- Edge AI appliances
- Sovereign infrastructure
- Multi-arch Docker builds
- ARM64 deployment validation
- React
- Next.js
- TypeScript
- Tailwind CSS
- Realtime interfaces
- WebSocket runtimes
- AI chat applications
- Mission timelines
- Operational dashboards
I have designed and operated software platforms serving hundreds of thousands of requests per day, including Go services handling around 800,000 daily requests with high availability.
My experience includes:
- Leading engineering teams
- Designing high-volume APIs
- Building Kubernetes-based platforms
- Developing secure healthcare applications
- Working with telehealth and healthcare messaging
- Creating AI and knowledge-management products
- Building private enterprise RAG systems
- Designing multi-tenant SaaS architectures
- Implementing authentication, authorization, billing and usage metering
I have worked on or am actively exploring AI systems for:
- Healthcare
- Pharmacy
- Accounting
- Energy
- Smart buildings
- Maintenance
- Legal and enterprise knowledge
- Research and education
- Public administration
I am interested in collaborations around:
- Agentic AI
- Private enterprise AI
- Edge AI
- Local LLM deployment
- NVIDIA DGX Spark
- Knowledge systems
- Industrial AI
- Healthcare and regulated industries
- Distributed systems
Email: [email protected]