I build AI tools that do real work. Not chatbot demos that hallucinate — production systems with guardrails, rate limiting, context management, and domain-specific knowledge. Every bot I deploy has been tested against adversarial inputs and edge cases.
I work across the major LLM providers: Claude for complex reasoning and code generation, Groq for blazing-fast inference, Gemini for research and content, and Ollama for free local AI. My MCP (Model Context Protocol) servers let Claude Code access external tools and APIs seamlessly.
AI assistants for your website with domain knowledge, rate limiting, and session management. Bilingual support.
Multi-tool AI agents with web fetch, code execution, file reading, and conversational memory.
Custom Model Context Protocol servers that extend Claude Code with external APIs, databases, and tools.
Content management systems with Claude API integration for automated content generation and design management.
Three custom Model Context Protocol servers: Gemini (research/content), Groq (fast inference), and Ollama (local AI). 27 tools total.
AI-powered Discord bot with TypeScript, discord.js, and Groq Llama 4 Scout. Multi-round agent loops with tool use.
AI-powered CMS platforms with Claude API. Automated content generation, design variant management, and gallery deployment.
Let's discuss your project requirements. I'm based in the USA and Colombia, available for both local and remote work.
Get in Touch