Services
AI Engineering
AI agents, LLM systems, RAG, MCP and AI automation — built to run in production, not as demos.
AI Engineering
AI agents, LLM systems, RAG, MCP and AI automation — built to run in production, not as demos.
I design and build AI systems that do real work: what the model decides, what the code decides, and where the memory lives.
What this looks like in practice
- Memory for agents. I built CaBrain, memory infrastructure for AI agents: persistent memory, hybrid retrieval and an entity graph, exposed to any agent over MCP.
- Agents in production. At ID8 Media I delivered an Arabic-language AI intelligence platform: Go agents with multiple analyst personas producing Arabic decision briefs with confidence tagging, on Gemini and Cloud Run.
- Agentic tooling. Orchestra MCP is plugin-based infrastructure for agentic developer environments, and this site's own MCP server exposes its issues, notes and content to AI clients.
- An agentic operating layer. At One Studio I designed the agents, evals, guardrails, memory, observability and decision rules that let ventures run with fewer people.
Typical engagements
Designing an agent architecture, adding retrieval or long-term memory to an existing product, exposing a product to AI clients over MCP, or taking an AI prototype to something that survives real traffic.
What is included6
- AI agents
- Agent memory
- RAG
- MCP servers and tools
- LLM integration
- AI automation