Team A2: ServeSense

Carnegie Mellon ECE Capstone, Spring 2026: Mario Cruz, David Hershenson, Geronimo Carom

Tag: status report

Geronimo’s Status Report 4/25
Posted by : gcarom Geronimo's Status Reports, Status Report

This week I made some great progress on adding some new aspects to the app to help amplify the user experience. I added gamifying components that include, challenges, progress trackers, and ways to compare how you are doing compared to other users. I also incorporated an LLM that gives you feedback on specific aspects of your game. I also was able to incorporate a new aspect to the ML Model and now it can also classify differences in levels of players. This was mostly for proof of concept so that later on we can categorize players based on their UTR.

One thing that I need to continue to do is collect data of more swings, and different types. I also need to continue working on the swing visualization and add a couple of components to it.

Currently I do not have an blockers.

Team Status Report 4/18
Posted by : gcarom Status Report, Team Status Reports ,

Our team made strong progress this week in many aspects of the project. First, we improved the hardware design for the racket attachment. The new prototype is smaller and fits the key components better than the old version. We still need to make more progress on the PCB, so that’s one thing we are working on. Second, we made great progress on data collection and the ML pipeline. We collected a large amount of swing data from many players, improved the bluetooth flow to make collection more reliable, and built a model that can classify stroke types with high accuracy. We also continued working on swing visualization, and we still need to improve its accuracy. Third, we improved the app experience by adding an AI coach and a profile section that gives users summaries and feedback on their progress.

In order to mitigate risk 1, we will keep iterating on the hardware design so it fits well on the racket and does not negatively effect the user experience. We will also keep working on a final version of the PCB and overall attachment design.

In order to mitigate risk 2, we will continue collecting more data, improving the model, and testing the accuracy. This will help us incorporate more swing types and improve consistency across users.

In order to mitigate risk 3, we will keep working on the swing visualization and continue checking the reliability of our firmware, bluetooth transmission, and sensor data. We want the full system to stay stable during real use.

We also made progress outside the core prototype. We prepared for an upcoming pitch competition by working on the BOM, user guide, market research, unit economics, logo design, and presentation script. Overall, the team had a productive week and we do not have any large blockers right now, but our main focus next week will be final hardware design, better visualization, more data collection, and further improvement of the ML model and AI coach.

Geronimo Status Report 04/18
Posted by : gcarom Geronimo's Status Reports, Status Report

This week I was able to make a few good strides on the project. I was able to gather data with my team and then create an ML model that would classify stroke types correct >95% of the time. I also incorporated an AI coach that gives users feedback based on their focuses and their sessions. I also added a My Profile section to the app that gives the player summaries on what they have done and how they are improving.

There are a few things I was not able to get to such as more types of swings in classification.

There are currently no blockers.

My goal is to continue collecting data, improving the model, and improving the AI coach

Team Status Report April 4
Posted by : gcarom Status Report, Team Status Reports ,

Our team is making steady progress on all fronts with regards to calibration testing of the IMU, the design to hold the IMU, and visualization of data. Our biggest priority now is data collection for fine tuning a model that will categorize different swings. This will allow us to make large strides in our project.

To accomplish this goal properly we need to both collect enough data and collect data from a variety of player levels. Once this is done we will be able to then fine tune the model and test its accuracy.

After this is complete we will also be testing the device during a proper session to see how it holds up and making sure the data for sessions is being organized and displayed properly.

Geronimo Carom Status Report April 4
Posted by : gcarom Geronimo's Status Reports, Status Report

This past week I worked closely with Mario to improve the swing path generation. We made great progress and the swings to be generating consistently with some reoccurring issues that are being debugged. I also did some research on how we would be planning on training the model for the project and found a pre-trained model we can use that interprets IMU data. All we need to feed it is labeled data from out IMU

Progress this week was good but we unfortunately were not able to record some data. We worked on making sure the data incoming from the IMU was accurate to make sure the data we were going to feed the model is correct.

We’re a few days behind on data collection because varsity tennis reserved the courts this weekend for a weekend tournament they had.

For this coming week I am looking to collect some data and start the training process of categorizing swings

Geronimo’s Status Report 03/28
Posted by : gcarom Geronimo's Status Reports, Status Report

This week I worked on connecting the IMU to the app through bluetooth. The first step was to upload the app on my phone and not work on xcode anymore. After that I had to work with Mario to connect the IMU via bluetooth and be able to read the data. Currently we have base functionality when connecting and reading data.

Progress this week was steady and we focused on preparing for the demo.

My goal for this upcoming week is to keep working on the device reading of sessions and also some work with the type of swing reading.

I haven’t started the training for determining types of swings yet which puts me a bit behind, but I am sure I will be able to catch up on this.

Geronimo’s Status Report
Posted by : gcarom Geronimo's Status Reports, Status Report

This week I was able to integrate the data being collected from the IMU to the app and have it produce a swing path. This still needs further testing and refining but the ingestion of data seems to be working.

For the coming week I hope to continue testing the swing path interpretation and start work on the other aspects: predicted speed path, type of serve, swing path labeling, etc.

No blockers for this week

Geronimo’s Status Report 3/14
Posted by : gcarom Geronimo's Status Reports, Status Report

I spent this week continuing general work on the UI of the app and some general clean up so that the data we receive will be better organized and easy to work with. Towards the end of the week we got a JSON of some fake data so I started integrating that into the app so that it can read the JSON’s properly.

I am currently on schedule, but I want to start really pushing fro data collection to be a priority

For next week I hope to do at least one session of live testing after integrating the app.

Team Status Report 03/07
Posted by : gcarom Team Status Reports ,

Our team continues to make steady progress toward our MVP and currently sees no major risks to completion. This week we continued progress on hardware development, embedded systems, and the application. We had a couple iterations on the hardware attachment that attaches the device to the tennis racket. On the embedded side, we updated the controller code to use SPI instead of I2C and successfully read IMU data packets at 500 Hz while transmitting data via BLE. On the software side, we began integrating the IMU data stream into the app and started implementing calculations to approximate ball velocity.

One potential challenge that carried on from last week is finalizing the physical hardware design. The attachment must be securely mounted to the racket while avoiding movement during play. When we begin collecting real swing data, we may also need to adjust parts of the data pipeline or processing steps to ensure the data is reliable and can be used for our specific use cases consistently.

To mitigate hardware risk, we will continue iterating on the attachment design and finalize the prototype so we can begin collecting swing data. We will also finalize component selection and place an order for a PCB.

To mitigate data and ML risks, we first need to establish a working end-to-end pipeline before focusing on model accuracy. We will collect sensor data via BLE, run basic analytics on the laptop, and begin preparing the dataset for model training. We will also continue adapting the app so it can interpret the incoming IMU data and compute metrics accordingly.

We do not have an updated schedule.

Our focus next week will be continuing to work on the physical prototype, collecting initial swing data, integrating the IMU data pipeline into the app, and beginning early model training for stroke classification.

Geronimo’s Status Report 03/07
Posted by : gcarom Geronimo's Status Reports

This week I have begun adapting to the app to be able to use and interpret the data being streamed in from the device according to the format of the data. I have also begun working on the calculations for the approximated ball velocity. And finally have started to think about how we should be storing the users data, locally and a little at a time in compressed files, or in an AWS Bucket.

Progress seems to continue to be steady.

My goal for this coming week is to continue the integration with the IMU and calculations of ball speed so that next week I can focus on training the model to interpret different types of swings. I do have to leave early this week to New York so I am hoping that does not set me back.