Point

AE Studio designed and shipped a publicly available iOS app in 12 weeks. This small footprint allowed for fast decision-making and tight iteration cycles.

The project was structured in four concurrent phases: app development, expert knowledge gathering, data streaming infrastructure, and the health recommendation engine. Running these in parallel rather than sequentially was the key to hitting the timeline without sacrificing quality.

The recommendation engine was built in close collaboration with Point's team of exercise scientists and medical advisors. Rather than relying on generic fitness heuristics, we interviewed the domain experts to understand which health indicators mattered most, how they interacted, and what recommendations were appropriate for different user segments.

The result was a proprietary health scoring system that combined rule-based logic with machine learning models. Rules derived from expert knowledge handled known patterns. ML models were layered in to capture more nuanced signals as user data accumulated over time. This hybrid approach let the system deliver value immediately while improving continuously.

The app connected to users' wearable data through Apple HealthKit, bringing in workouts, movement, sleep, and biometric readings into a unified view. The data pipeline was built to clean and normalize raw sensor data before feeding it into the recommendation engine, handling the noise and inconsistencies inherent in consumer wearable hardware.

Real-time data processing ran on AWS Aurora, handling complex health metrics within milliseconds. This architecture was chosen for its scalability, ensuring the system could grow with Point's user base without requiring major re-engineering.

The core product experience was not a dashboard. It was a recommendations feed. Rather than presenting users with raw numbers, the app told them what to do next: whether to push harder in their next workout, prioritize recovery, adjust sleep habits, or focus on a specific aspect of their training.

Recommendations were personalized based on user goals, survey responses, wearable data, and individual health scoring. Users could set goals and track progress toward them. The system also detected when rest or recovery was warranted, helping users avoid overtraining rather than just cheering them on.

Keeping the Point team in control of their own product was a design priority throughout the engagement. We built editable no-code admin tools, enabling Point's product team to create and update insights and recommendations without requiring engineering support.

This gave Point operational independence and allowed them to iterate on the product content layer quickly, responding to user feedback and evolving their health content without bottlenecks.

After the initial app was live and gaining traction, AE Studio recommended a strategic pivot: rather than positioning Point purely as a consumer app, the underlying health engine could be packaged as an SDK and licensed to other companies that wanted to embed personalized fitness insights into their own products.

This shift expanded Point's addressable market significantly. AE Studio built out the SDK, creating a standalone product that integrated biometric data from wearable devices, normalized and cleaned the data, and surfaced real-time physiological insights and personalized recommendations for third-party integration.

The SDK pivot helped Point secure follow-on funding and positioned the company for the acquisition that followed.

Beyond the initial MVP, AE Studio served as Point's complete engineering organization. This meant not just building features but making architectural decisions, managing infrastructure, and evolving the product strategy alongside the founding team.

Amplitude was used for analytics and engagement tracking, giving Point visibility into how users were interacting with the app and where the recommendation engine was driving behavior change. This data informed ongoing product decisions throughout the engagement.