What is PFP? Personalised Feeding Plan (PFP) is a smart feeding algorithm to help pets reach their target weight while on the ilume tech suite. It uses a combination of activity tracking, feeding and timely body measurements to ensure that the dog reaches their target weight or maintains their ideal weight once reached. Responsive Feeding However, we needed our daily calorie requirements to be dynamic so that we adjust meals to over or under compensate for the day’s activity. For example, if the dog usually don’t go for a big walk Sunday morning, but this Sunday if he does. By dinner time, we need to increase the feeding while also making sure the dog reaches their target weight.

Give the user a prompt to confirm that the following data looks correct: Walks - Start and End Times Feeding - Remaining Food in bowl Recovery - Bedtime and Wakeup time Activity Type - Zoomie, Running & Walking values The backend will store these validated values when training the ML model.

The objective of this task is to develop a feature that allows a tracker device to automatically sync data with a user's phone when it is out of range of its primary hub. This functionality ensures continuous data collection and accessibility even when the tracker cannot directly connect to the hub.

The objective of this task is to develop a feature that utilises the ChatGPT API to allow users to input a food item and, based on the breed of their dog, determine if the food is safe for consumption.

This feature aims to develop an achievements engine to track and manage user achievements and support rolling statistics for continuous user engagement.

The goal of this task is to develop a comprehensive and secure Over-the-Air (OTA) update system that enables remote software or firmware updates on devices without requiring physical access. The implementation will ensure a seamless and user-friendly experience

The aim of this project is to implement logging for the hub and devices using AWS IoT services. Alternatively, if we choose not to use AWS IoT logging, we would need to implement our own logging system through the API. However, this alternative would take longer to implement, and there is a risk that our own system will likely not perform as well as Amazon's AWS IoT logging service. Additionally, implementing our own system may increase log streaming traffic, resulting in additional costs and potential outages.

The goal is to transition our syncing mechanism from HTTP to MQTT. This change aims to enhance the efficiency and reliability of our data synchronisation process, especially in environments where bandwidth is limited, network stability is a concern, or real-time updates are crucial

Implement push notifications that allow targeted and personalised notifications to be sent to users. This feature aims to increase user engagement by delivering timely and relevant updates directly to users' devices

Develop a feature flag system that allows for gradual rollouts and targeted user access to new features. This will enable controlled deployment, testing, and customisation of features to specific user segments.

Optimise search functionality using Algolia, prioritising closest matches, incorporating brand, protein, and name, and implementing fuzzy search for improved user experience. Closest Match: Implement Algolia's indexing capabilities to ensure the closest match to the user's query appears at the top of search results. This involves leveraging Algolia's relevance ranking algorithms to prioritise results based on query string similarity, popularity, and relevance. Brand, Protein, Name: Structure the search algorithm to give the highest weightage to the brand, followed by protein, and then the product name in the search results. This entails configuring the indexing strategy to consider these attributes when ranking and displaying search results. By doing so, users can easily find products based on their preferred brand, protein requirements, or specific product. Fuzzy Search: Integrate fuzzy search functionality to account for typographical errors, misspellings, or variations in the search query. Algolia's fuzzy search feature enables the system to recognise and retrieve relevant results even if the user's input contains minor errors or deviations from the exact search terms. This enhances the search experience by accommodating user input variations and increasing the likelihood of finding desired items.

Implement dynamic thresholds that adjust based on breed and size to ensure more accurate data collection and analysis. This feature aims to provide tailored data insights and recommendations, enhancing the relevance and precision of information for different dog breeds and sizes.

Develop a feature that enables data visualisation to facilitate easier consumption and reveal insights and trends. This feature aims to enhance user understanding by presenting data in an accessible and visually appealing format.

Implement a feature that allows users to set up a baseline feeding schedule. This will enable a gradual decrease in food amount to ensure pets transition smoothly to a new feeding regime without experiencing hunger.

Implement a pull-to-sync feature that allows users to refresh and synchronise data within the app by pulling down on the screen.

Develop an onboarding flow that guides new users through creating a pet profile, completing their profile, and setting up devices. This feature ensures user engagement by providing a seamless initial experience.

When I view activity on Tuesday it shows monday.

Weekly report on your dogs activity levels