Activity

Jow
Jow worked on GemClassify
about 1 month ago
3h 20m logged

Built an end to end gemstone classification web app using FastAPI and TensorFlow, powered by a custom CNN trained to distinguish authentic and imitation gemstones. Designed and refined the full frontend experience, including a polished landing page, improved prediction workflow, responsive UI, batch upload support, confidence visualization, and cleaner interaction details like custom cursor and text selection styling.

Attachment
Attachment
Attachment
Attachment
Attachment
Attachment
0
Jow
Jow shipped face-recognition
about 1 month ago

Shipped this project!

I built a Simple Face Recognition system using Python and a KNN classifier to identify and classify faces from images.

This project taught me a lot about machine learning workflows, image processing, and model training pipelines. I also learned how to turn a machine learning model into a usable product by building a REST API with FastAPI and connecting it with simple web and client interfaces.

One of the biggest challenges was improving accuracy and handling different image conditions, so I added image enhancement and batch training features to make the model more reliable.
I also practiced deployment skills by Dockerizing the project and preparing it for cloud deployment.

Overall, this project helped me connect ML concepts with real-world application development, from training a model all the way to deploying a working system.

Jow
Jow worked on face-recognition
about 1 month ago
9h 35m logged

Identify and classify faces using trained KNN model

Attachment
0
Jow
Jow worked on face-recognition
about 2 months ago

I’m working on my first project! This is so exciting. I can’t wait to share more updates as I build.

Attachment
0