

Senior Applied AI Engineer and AI Solutions Architect with 23+ years of software engineering across interactive products, enterprise applications, and agentic systems. I implement AI at any layer and any scale: agentic enhancement of an existing product or a full system from scratch, cloud or fully local, from hardware-level inference optimization through multi-agent orchestration to the application on top.
Proven delivery of production AI: agentic workflows and multi-agent architectures, MCP and API integrations across HTTP, SSE, and WebSocket transports, RAG, computer vision, multimodal generative AI across video, image, speech, and audio, and web3 tooling. End-to-end ownership from business requirements to deployment, in Agile cycles with stakeholders in the loop; architecture, integration tests, and security review stay human-in-the-loop.
1st place, Anna AI-Native App Hackathon 2026. Current work is public and verifiable: a live self-publishing media portal in production, more than ten published agent skills and toolkits, and open-source inference optimization on AMD hardware.
Public repositories on GitHub
Agentic Engineering
Multi-Agent Systems
AI Agents
MCP (Model Context Protocol)
Context Engineering
Prompt Engineering
RAG
Generative AI
Computer Vision
Local LLM Inference (vLLM, ROCm)
Python
Docker
TypeScript
REST, SSE and WebSocket APIs
1st place, Anna AI-Native App Hackathon 2026, with gamentic: a local multi-agent AI dungeon RPG, one agent per character, world state behind 40 validated tools.
El Censurado, a live self-publishing news portal in production: a full agentic editorial workflow, 32 tools driven through a 14-step gated walk.
Published agent skills and toolkits: text-to-3D, OCR, research, web search, Telegram, and blockchain.
Open-source local inference on AMD Strix Halo: a custom HIP kernel, AWQ-INT4 at 256K context, and published benchmarks.