Footprint: company reports built through AI workflows.
A platform for configuring multi-agent workflows through a visual interface. Each agent has a dedicated model and tool set and can run in parallel or in sequence. The resulting company report is assembled, reviewed, and translated when needed.
The project, at a glance
The request specifies the company, sector, and questions the report needs to address.
The interface defines which tasks run in parallel and which depend on the output of an earlier step.
Each agent has a defined task, model, and tool set. Separate steps collect business data, analyze the sector, and gather public signals.
An assembly step combines the outputs into a single draft.
A review step checks the draft for missing information and internal inconsistencies. Translation follows when required.
The final document retains a traceable record of the workflow and its dependencies.
The workflow is assembled in the interface, step by step.
Independent tasks run in parallel; dependent tasks wait for the preceding output. Each step uses the model and tools selected for its requirements.
Workflows can be configured through the interface or the API, and new steps can be added without rewriting the rest of the pipeline.
Users can adjust workflow composition without requesting a code change for every update.