top of page
Sift Logo.png

An open-source repository for forest restoration practitioners by leveraging AI to collate resources to enable efficient project planning.

Built for Collaborative Earth â†—︎

Tree pattern.png
Sift Thumbnail.png

Collaborative Earth

As part of our capstone project, we partnered with Collaborative Earth, a non-profit organization dedicated to restoring biodiversity worldwide.

We worked with the Assisted Forest Regeneration Lab whose mission is to support global reforestation goals with a tool to gather and synthesize available reforestation knowledge (including unpublished data from field partners, and non-English sources).

Roles

UX Research
Design Systems
Usability Testing

Team

Anusha Thalnerkar
Kirtana Kannan
Priyam Shah
Rishma Bora

Duration

7 months

Nature of the Project

A project embodying the true collaborative spirit

Collaborative Earth

We partnered with the Assisted Forest Regeneration Lab.

California College of the Arts

I worked with a diverse team bringing expertise in graphic design, research, and business acumen.

Forest Restoration Community

We conducted co-design workshops, & usability tests with 15+ practitioners.

Theory of Change Framework

This framework acted as a guiding anchor for our project, consistently keeping us aligned with the right vision.

Problem Area.png

Forest restoration practitioners face an overwhelming task of sifting through sparse information.

Problem Area.

Solution.png
Sift Logo.png

We designed a repository which facilitates planning & collaboration, by leveraging
AI to collate resources.

Designed Solution.

Forests.png

By emphasizing accurate decision-making, we can facilitate inter-lab collaboration for improved restoration outcomes.

Envisioned Impact.

With the niche domain
of the project, we faced
a few 
challenges.

Tree pattern.png

Challenge 01

Sense making of complex data and organization of information to build a product from scratch.

Especially when all we started with was an Excel sheet, which required us to make sense of the data and develop an entirely new tool.

Excel Sheet.png
Tree pattern.png

Challenge 02

Striking the balance between designing an intuitive yet creatively unique product.

In order to build a tool that would help easily access and organize vetted knowledge, we had to think of functionality first.

Analogous Research.png

Solution Overview: Key Features

Sift Logo.png

An open-source repository for
forest restoration knowledge.

1. Dynamic Search 

Dual search functionality enabling forest restoration practitioners to search and get quick insights on documents they upload.

2. Seamless Integration of AI

Smart Insights.jpg

Trained on a domain specific Large Language Model with thousands of vetted resources, Sift generates succinct AI summaries. 

3. Workspace for Practitioners

My Archive.jpg
Collections.jpg

Making collaboration easier by introducing workspaces for practitioners that are working in silos can feel more connected and upto date with knowledge.

User Journey of a Forest Restoration Practitioner

Professionals dedicated to restoring and rehabilitating degraded forest ecosystems to their natural, healthy state.

Chosen area of focus: Researching and Planning

Effective project planning leads to successful outcomes, which when reported, assist other practitioners in planning their own projects.

Diving Deep: Research to Insights

Through conversations with 15 restoration ecologist, org leaders and practitioners, we learned about their needs & challenges.

Theme 1.jpg

Theme 2

Variation in Tools

“I would note that use of a mapping program has been a bit of an obstacle maybe in certain projects.”

Organization Lead at Collaborative Earth

Theme 2_edited.jpg

Theme 1

Sparse and scattered documentation

“Sometimes we end up with over 20,000 abstracts, and then have to sort through the relevant ones to get the necessary data.”

Restoration Ecologist at Collaborative Earth

Theme 3_edited.jpg

Theme 3

Drivers of success: involvement of community and stakeholders

“For restoration to be successful, it should improve people’s livelihoods - primarily financial.”

Restoration Ecologist at Collaborative Earth

Our Process Principles

Our process was non-linear, & extremely iterative.

Laptop Outline.png

How was it possible to keep iterating on our prototypes?

Building a Design System

Design Systems
Card.png
Iconography.png

Detailed Solution

Onboarding the
Practitioners

When forest restoration practitioners share essential details about their work and specific projects, it enables Sift to create relevant and valuable content.

Searching for
resources

Given the unique nature of restoration efforts in each locality, Sift offers a robust filter panel to help practitioners access the most precise information.

Accurate Insights

Smart Insights using AI are generated taking into account three scenarios: 
Summary of the search result, an individual resource and summary extracted from a document uploaded or based on prompt.

What didn't work:
A standalone ChatGPT-like UI was removed after understanding practitioners' mental models - a seamless AI integration was needed to build trust.

Practitioner's 
Workspace

The workspace enables collaboration with a Personal Archive for organizing resources and Collections for teams to share and communicate through notes.

What didn't work:
Project documentation was considered, but a dedicated space for resource collection was more viable given the abundance of existing platforms.

Measuring Impact

The tool is currently in development, with the dataset still being trained. Here are some of the anticipated impacts.

User Engagement

Total user base
Rate of new users logging in
Geographic Distribution

Number of
partnered labs

Quantifying the network, and building new collaboration ways.

Content Utilization

Number of resources saved
Number of projects

Tree pattern.png

Learnings & Reflections

When in doubt,
test it out!

Every page and feature was finalized through community feedback and multiple iterations to achieve our desired outcome.

Working 
with AI

Exploring the potential and limitations of AI; 
Understanding people's perceptions of it.

Being able to pivot, and realigning focus.

We often had to pivot from initial concepts, with our 'Theory of Change' framework serving as a guiding anchor.

bottom of page