
PROJECT OURANOS
How might we accurately detect environmental patterns faster, so conservation efforts can have maximum impact?
NDA
This case study represents a synthetic context, not the project's actual deliverables. It's adapted to focus on ecological system monitoring.
Analysts can spend dozens of hours manually analyzing data. Project Ouranos was one of the earliest adopters of AI in this domain by making sense of information from many sources (i.e., LiDAR, satellite) and many formats (i.e., still, video), so analysts can act on environmental threats faster and more safely.
SERVICES
-
UX Research
-
UI Design
CLIENT
-
US Federal Government, Executive Branch
Impact
With our high-velocity Design Sprint, we acheived:
Estimated labor savings of 400 hours (~$100k) when compared to conventional design processes
Quick leadership buy-in and multi-stakeholder alignment on goals and objectives

FIVE-DAY DESIGN SPRINT
Day 1 - Map
The first day established the strategic framework for our sprint week. Our primary objectives were to:
-
Define a clear long-term vision
-
Map the core challenge and key stakeholders
-
Develop actionable "How might we?" problem statements
Meet Our Users!
THE OPERATOR
Top 3 goals:
Identify environmental hazards
Protect false-positive patterns from action
Escalate appropriate findings to Analysts
Top 3 needs:
-
Maintain precise and granular control over applications to meet specific task goals
-
Review results of the job that identified particular environmental patterns
-
Quickly take action to escalate potential hazards to Analysts
THE ANALYST
Top 3 goals:
Identify the risk level of an Operator's findings
Protect false-positive findings from action
Escalate true hazards to leadership
Top 3 needs:
-
Receive timely alerts about possible environmental hazards
-
Read clear, precise, and complete information about the identified hazards
-
Quickly take action to escalate the potential hazards to leadership
THE DEVELOPER
Top 3 goals:
-
Integrate new/updated data sources into the data lake
-
Create interfaces that allow Operators and Analysts to successfully complete their tasks
-
Continually enhance the system to optimize the value of the platform
Top 3 needs:
-
Maintain clear, understandable code frameworks for maximum development efficiency
-
Develop rigorous documentation of system architecture and processes
-
Create scalable, repeatable patterns for new features and enhancements.
Day 2 - Sketch
In Day 2, we shifted focus to solutions, identifying opportunities for improvement and freely brainstorming solutions through rapid sketching.
Day 3 - Design
We evaluated each sketch from yesterday to choose the concept most aligned with our objectives, then integrated the key elements from the winning sketches into prototype plans. We voted collectively on the ideas we would actually prototype the next day.
Day 4 - Prototype
We developed the top-voted features from the previous day into a functional prototype aligned with our key objectives, prioritizing usability and efficiency. Given the need for complex workflows — which demands a wide field of view for multitasking and task management — we adopted a desktop-first approach with a parallel mobile experience to maintain seamless cross-device functionality.
A prototype of the Project Ouranos homepage, which prioritizes high-value common tasks for the analyst.
Day 5 - Test
After building and demonstrating the prototype, we conducted user testing sessions and customer interviews, carefully observing their interactions and reactions. By day's end, we had clear insights into both our progress and the necessary next steps.
What happened next?
After our 5-day Design Sprint, we presented our prototype and findings report to executives. This delivery represented a cross-functional collaboration to demonstrate how user experience improvements directly supported both business objectives and revenue potential.
Our solution introduced an innovative approach: leveraging machine learning to synthesize disparate data sources for real-time environmental hazard detection. The initiative demonstrated such significant potential that it evolved into a formal proof of concept later that year, serving as the foundation for advanced AI projects utilizing multi-source data integration.
Reflections
By adopting a focused high-veloicty Design Sprint approach, we delivered measurable outcomes in just one week -- avoiding the drawn-out timelines (often two weeks or more) typically consumed by rescheduled meetings, fragmented workshops, and isolated work.
Our team embraced agile principles at speed, thus enabling rapid alignment and execution, proving the power of early and iterative goal-setting.















