Software architect and AI engineer with over a decade of building production systems and four years deep in AI/ML. I design and build intelligent systems, from multi-agent orchestration and LLM fine-tuning to full-stack AI-native applications, cloud infrastructure, and cybersecurity.

My current work is focused on custom AI agent development using CrewAI, LangGraph, and Claude Agent SDK, fine-tuning open-source LLMs with LoRA/QLoRA, building RAG systems with vector databases, developing MCP servers for agent tool ecosystems, and AI security including prompt injection defense and red teaming.

On the engineering side, I work across full-stack web and mobile applications, REST API design, cloud-native architecture on GCP and AWS, DevOps pipelines, and IT consulting for digital transformation. I also do security audits, penetration testing, and compliance work.

Tech I work with: Python, PHP, JavaScript, TypeScript, React, LangChain, LangGraph, CrewAI, Hugging Face, PyTorch, PostgreSQL, MySQL, Redis, Pinecone, ChromaDB, pgvector, Docker, Linux, Caddy, Nginx, and Git.

I believe in building systems that are intelligent and not just automated, shipping production-ready code instead of prototypes, and keeping architecture clean for the long term. Security and safety in AI systems are not optional.