Insights

Cambridge engineer Ben Chang reflects on his summer internship at Aeris-UK, working across edge AI, real-world data and client-facing R&D, and how the experience shaped his plans for a career in AI.

Aeris-UK presented its work on agentic cyber defence at the land tactical edge and demonstrated the CEDAR prototype live at AMLUCS 2026.

Training AI models can be computationally expensive, particularly for smaller teams. This article examines practical ways to reduce energy use and training costs while maintaining effective model performance.