Applied research,
not academic exercises
Every research line at Tangison Labs starts with a problem we have encountered in production. We study, prototype, validate, and then ship. Nothing stays in the lab unless it works in the field.
Research areas
Applied AI
ActiveBuilding practical AI systems that work in production. Our research focuses on making intelligence useful: from workflow automation to natural language understanding, every model we train solves a specific problem. We prioritise reliability and interpretability over raw benchmarks.
Active projects
- Agent orchestration protocols
- Local-first inference patterns
- Context-aware task routing
Workflow Automation
ActiveDesigning systems that eliminate repetitive work without eliminating human judgement. Our automation research targets the gap between what AI can do and what people actually need done. Every automation we build preserves override control and maintains audit trails.
Active projects
- Skills-based site creation (Webman)
- Multi-agent coordination
- Pipeline validation frameworks
Infrastructure Systems
ExploringResearching the foundational layers that make AI products viable at scale. From deployment pipelines to observability stacks, we study how to run intelligent systems reliably in real-world conditions. Our infrastructure work feeds directly into Studio and Agent.
Active projects
- Edge deployment patterns
- Observability for AI systems
- Cost-aware scaling strategies