Discover the future of freight quoting with our automated system. Designed to enhance efficiency and accessibility, our system streamlines the process, eliminating the need for lengthy phone calls and complex navigation.

Project Overview

What is a Freight Quote?
A freight quote is an estimated price for shipments, determined by factors such as distance, weight, date, and any special services required.

Problem:
Many business owners, especially those with less technical literacy, face challenges in obtaining freight quotes instantly. They often have to endure lengthy phone calls or navigate complex online platforms.

Impact:

• Increased Efficiency: Our system eliminates the need for lengthy phone calls and complex navigation, providing instant quotes.

• Enhanced Accessibility: By minimizing digital inequality, we level the playing field for all business owners, ensuring they can easily access the information they need.

Our Solution

We developed a Gmail control system that automatically extracts incoming emails from users and uses an AI model to extract relevant information. This information is then used to provide customers with accurate and instant freight quotes.

Key Features:

• Email Integration: Seamlessly send and receive quotes via Gmail.

• Advanced Information Extraction: Utilize AI models to accurately extract key information from emails.

• Instant Quotes: Use the Zuum API to calculate and deliver freight quotes directly to users’ inboxes.

Challenges

Learning and Implementing AI:

• Our team faced the dual challenge of learning and implementing AI technologies simultaneously. This required additional effort and caused delays in the initial stages of the project.

Data Dependency:

• The performance of our AI model heavily relies on comprehensive datasets. Incomplete or insufficient data can hinder the model’s effectiveness and accuracy.

Model Flexibility:

• Ensuring that our AI model is adaptable and can be easily fine-tuned to handle various cases is crucial for the system’s long-term success.

Embrace Fallibility of AI

Quick Response to Model Flaws:

• We are committed to reacting quickly to any flaws or inaccuracies in our AI model. Continuous monitoring and rapid iteration are key to maintaining model reliability.

Streamlined Data Aggregation:

• To improve model performance, we streamlined the process of aggregating datasets, ensuring that data collection, testing, and fine-tuning are integrated and efficient.

Unified Testing and Annotation:

• We established a seamless workflow where testing, annotating, and fine-tuning the model happen in one place, allowing for faster and more effective updates.