Thoughtworks Technology Radar 33 Highlights AI Assistance and Context Engineering

Thoughtworks has released Volume 33 of its Technology Radar, which focuses on the rapid evolution of AI assistance in enterprise software, announced in a press release. The biannual report, informed by client experiences, outlines key shifts in how organizations deploy and manage AI systems.

This edition emphasizes the growing importance of 'context engineering', the adoption of the Model Context Protocol (MCP), and the expansion of agentic systems that enable semi-autonomous workflows. Compared to the previous volume, which highlighted retrieval-augmented generation and prompt engineering, the new report signals a move toward more mature and integrated AI practices.

The Radar also notes a rise in infrastructure orchestration for AI as companies manage larger GPU workloads for model training and inference. Additionally, it identifies emerging 'AI antipatterns', including shadow IT and overreliance on AI-generated code, emphasizing the need for continued human oversight and technical rigor in AI-driven development.

Chief Technology Officer Rachel Laycock stated that the industry has shifted from experimental 'vibe coding' approaches to more structured efforts addressing context, infrastructure, and security, reflecting a year of active experimentation and growing maturity in AI adoption.