James Trappett
James Trappett's AI research focuses on the practical applications of artificial intelligence and machine learning, with particular interest in large language models, autonomous systems, and AI-augmented software development.
Research into the capabilities and limitations of large language models including GPT, Claude, and open-source alternatives. Investigating prompt engineering, chain-of-thought reasoning, tool use, and autonomous agent architectures.
LLMsGPTClaudePrompt EngineeringExploring how AI can transform the software development lifecycle — from code generation and review to automated testing, deployment, and infrastructure management. This website itself is a case study in AI-driven development.
AI CodingAutomationDevOpsCode GenerationResearch into AI systems that can plan, execute, and iterate on complex multi-step tasks with minimal human intervention. Investigating tool use, memory systems, and self-correction mechanisms.
AI AgentsAutonomyPlanningTool UseUnderstanding the implications of increasingly capable AI systems. Research into alignment techniques, safety guardrails, and responsible AI deployment practices.
AI SafetyAlignmentEthicsResponsible AIJames Trappett conducts practical AI experiments to test theoretical concepts in real-world scenarios. This website is one such experiment — demonstrating that AI can autonomously create, deploy, and optimise a complete web presence.
See the blog for detailed write-ups of experiments and findings.