Juba
AI-Powered Investigative Knowledge Graph
Inside the project
01 / 01Juba
The brief
A problem worth
building around.
Visualizing complex relationships between entities (people, companies, events) requires intuitive graph rendering while maintaining performance with large datasets. Building an AI research agent that can autonomously discover connections adds another layer of complexity.
Our approach
Juba maps networks with an interactive force-directed graph. The Research Agent uses leading AI models with tool calling to discover entity connections on its own. Session fingerprinting and encryption protect sensitive investigations across multiple layers. The crime board aesthetic ties the pieces together with pinned cards and connecting threads.
The experience
What it lets
people do.
The capabilities that turn the underlying engineering into a usable product.
- 01
Interactive network visualization
- 02
Entity profiles and dossiers
- 03
Timeline event tracking
- 04
AI Research Agent with tool calling
- 05
Cross-entity search with categorization
- 06
Admin dashboard with full CRUD
- 07
Crime board dark UI aesthetic
- 08
Article AI generation
Project record
What came out of it.
Interactive force-directed graph visualization
AI Research Agent with autonomous tool calling
Entity profiles for persons, companies, events
Timeline view with chronological event tracking
Multi-layer session security and encryption
Under the hood
Investigative journalism platform with AI research agent and network visualization
Investigative journalism platform with AI research agent and network visualization
Facing Similar Challenges?
Every business is different, but the problems tend to rhyme. Get in touch and tell us about yours.