March Updates

tl; dr

• New integrations: Hello Clue, TrainAsOne, Tredict, Hammerhead, and TrainingPeaks posting of planned workouts allow for more comprehensive tracking of fitness data.
• Our in-house machine learning model for Freestyle Libre pro sensors significantly increases glucose-reading accuracy to close to 100%.
• We now retry rate-limited requests to Withings, ensuring users always receive data.

📈 New Integrations:

👉 What:

🚀 We're excited to announce several new integrations, making it easier than ever to get started with our fitness data aggregation platform.

• Hello Clue: Track your menstrual cycle and connect it to your fitness data.
• TrainAsOne: Personalized training plans based on your fitness level and goals.
• Tredict: Plan and analyze your endurance training and racing.
• Hammerhead (Posting planned workouts): Seamlessly integrate your planned workouts from Hammerhead Karoo to your fitness apps.
• TrainingPeaks (Posting planned workouts): Streamline your training plan by posting your workouts from TrainingPeaks to your other fitness apps.

👉 Why:

By adding new integrations to our platform, we're making it easier for users to connect their data from different sources in one place.

👉 How:

We're constantly working to expand the number of integrations we support, making it easier for users to get started with our platform. Our goal is to provide a seamless user experience, regardless of which fitness apps and devices you use.

🧠 FreestyleLibre Pro calibration algorithm

📉 What:

Our team has developed an in-house calibration algorithm for Freestyle Libre Pro sensors, bringing our glucose-reading accuracy up to close to 100%. Our algorithm is based on a machine learning model that maps raw sensor data to the user's real glucose levels, improving accuracy and removing erratic values.

📉 Why:

Glucose-reading accuracy is essential for people managing their blood sugar levels, and we want to ensure our platform provides the most accurate data possible. Our customers want the best experience possible for their end users, and we strive to provide that.

📉 How:

We collected a large dataset of glucose readings and sensor data from Freestyle Libre sensors and used this data to train our machine-learning model. Our model now provides highly accurate glucose readings, which can be used to improve the management of diabetes and other blood sugar-related conditions.

🔄 What:

We're now retrying rate-limited requests for Withings, so you don't have to worry about handling rate limits imposed by providers. If a request to Withings for data gets rate limited, we'll automatically retry it at a later time when Withings rate limits are reset.

🔄 Why:

Rate limits can cause frustration and inconvenience for users, and we want to ensure our platform delivers data reliably and without interruption, taking pain points away from the experience rather than relaying them through

🔄 How:

Our platform keeps track of data requests, and when rate limits are hit, we automatically dispatch a task to retry the request at a later time when the rate limits are reset. This ensures that our users receive data, no matter what, and in a timely manner.