AI & research
Explore what’s possible.
Engineer what’s useful.
Our AI work connects language models with the realities of data, software delivery and operational support. Research feeds the products; delivery gives the research a practical purpose.
Research directions
The questions
we’re working on.
Local & distributed AI
Local language models, retrieval and distributed execution across CPU and GPU resources.
Explore the workProduct engineering & researchGoverned AI & automation
Combining language-model reasoning with deterministic workflows, validation, permissions and operational evidence.
Explore the workResearch concept / Proposed developmentEvidenceFlow
A Rosetta research direction for traceable, AI-assisted evidence briefings with source provenance and human review.
Explore the workApplied in Rosetta
From experimentation
into the workspace.
AI Chat, contextual support, semantic data access and generation tools bring the research closer to everyday work.
Our approach is to evaluate a model-backed feature within the system around it: its data, access controls, failure modes and the people who have to trust the result.
Explore Rosetta’s AI work
Collaborate with us
Bring a useful
research question.
We welcome conversations with software companies, research teams and organisations exploring practical applications of AI.
A good starting point is a specific task, representative data and a way to judge whether the result is useful. From there we can shape a proof of concept, evaluation or integration project.
Discuss AI & research