Tricentis buys AI-coding platform Tabnine to power its testing agents

Tricentis has acquired Tabnine , an AI-coding platform built for secure, context-aware enterprise software development. The Austin-based company will fold Tabnine's Enterprise Context Engine into its Agentic Quality Engineering Platform to sharpen the quality and testing agents it sells to large enterprises. Terms were not disclosed.

Rather than relying on standard retrieval methods, its Enterprise Context Engine builds a continuously updated knowledge graph of an organisation's systems. It maps entities, relationships, dependencies, and architectural patterns across repositories, documentation, tickets, APIs, and infrastructure metadata.

AI agents cannot reliably test software they do not understand, and shallow context can lead them to make inaccurate decisions and introduce risk. Tricentis argues the engine gives its testing agents enterprise-wide understanding to spot risk and speed up delivery.

"Quality engineering in the enterprise has never been a model problem. It has always been a context problem," said Kevin Thompson, chief executive officer of Tricentis. He said the context layer "belongs at the center of how we deliver software quality."

Organisations using the engine have reported up to a two-times improvement in AI accuracy, up to 80% less token consumption, and up to 50% faster resolution on complex tasks. Tricentis says those gains mean fewer false positives, fewer missed defects, and faster test cycles.

The combined platform offers dependency and impact analysis, automated governance against coding standards, and shared memory for multi-agent workflows. It can deploy on-premises, in a private VPC, or fully air-gapped to meet security and compliance needs.

The deal reflects a broader shift in enterprise AI, where reliability and deep context — not raw model power — are becoming the competitive edge. As autonomous agents move into mission-critical workflows, vendors are racing to give them the organisational knowledge to act without introducing new risk.