RoadRelay | Devpost

Inspiration

Did you know that using your phone while driving makes you up to 240% more likely to get into a car crash? Whether you're driving 20 mph through a neighborhood or pushing highway speeds, mobile notifications never stop, and neither does the temptation to check them, especially for phone-dependent young adults like us.

Apple CarPlay attempts to reduce distracted driving by silencing or limiting notifications, but in practice, this often creates the opposite effect. When drivers feel completely disconnected from their phones, many end up glancing at them anyway: at stoplights, during slowdowns, or even while moving, because they still want to know what’s important.

With RoadRelay, we set out to reduce the temptation to interact with phones while driving by offering something people actually want: safe awareness without interaction.

What it does

RoadRelay is a CarPlay-native app that lets drivers safely stay informed without touching their phone.

With a single tap in Apple CarPlay, RoadRelay:

Collects incoming notifications from the user's device
Prioritizes and summarizes them using Teli AI
Redacts sensitive content (to prevent information from being exposed aloud)
Reads a concise, driving-safe summary aloud through the car’s speakers
Sends the same summary as an SMS via Teli to be read aloud by Siri on CarPlay

Drivers can also ask follow-up questions verbally, such as whether anything urgent came in, and hear the response read aloud, all while keeping their eyes on the road and hands on the wheel.

Instead of forcing drivers to choose between silence and distraction, RoadRelay provides a safe alternative to illegal and dangerous manual phone checks.

How we built it

We built RoadRelay primarily using Flutter with Dart, allowing us to share logic while integrating deeply with iOS.

Key technologies include:

Flutter + Dart for the core app logic and UI
Native iOS (Xcode) for Apple CarPlay integration
MethodChannels to bridge CarPlay actions into Flutter workflows
Teli.ai’s API for AI summarization and SMS delivery
iOS Text-to-Speech for hands-free narration

When a CarPlay action is triggered, the native layer invokes a Flutter workflow that loads notifications, summarizes them, sends an SMS (to be read aloud by CarPlay), and updates the app state, all without requiring user interaction on the phone.

Desire Paths

This project fits the Desire Paths track because it directly addresses a gap between system design and real human behavior.

CarPlay is designed to mute notifications while driving, but drivers consistently ignore that design by glancing at their phones anyway. That behavior is the desire path: a signal that people don’t want zero information; they want controlled, high-level awareness.

RoadRelay embraces that reality and formalizes it into the product itself, offering a safer path that aligns with how people actually behave.

Challenges we ran into

Building a CarPlay app presented significant challenges due to Apple’s strict platform constraints. CarPlay apps are limited in UI complexity, background execution, and available APIs, which required careful design and creative solutions for features like voice interaction and narration.

We also faced challenges integrating Teli’s SMS API early on. Initial attempts failed to deliver messages correctly, but after working closely with the Teli.ai team and refining our API usage, we successfully implemented reliable SMS delivery.

Navigating these constraints while maintaining a smooth, safe user experience was one of the most difficult and rewarding parts of the project.

Accomplishments that we're proud of

Building a fully functional CarPlay app, a platform many developers avoid due to complexity
Creating a real, end-to-end AI workflow triggered entirely from CarPlay
Delivering a solution that addresses a real safety issue, not just a technical challenge
Designing around human behavior rather than forcing users into unrealistic patterns

Most importantly, we’re proud that RoadRelay tackles a problem that truly matters. A problem that has the potential to reduce distracted driving and save lives.

What we learned

Through RoadRelay, we learned the importance of building for platforms and use cases that are often overlooked. While developing for Apple CarPlay was challenging, it reinforced how impactful thoughtful design can be in constrained environments.

We also gained valuable experience integrating AI into real-time, safety-critical workflows and designing systems that prioritize clarity, restraint, and user trust.