stev3

Thank you. Will do.

Thank you! All native with tflite models trained on hand labelled data using pytorch. I do plan to open source everything if there's enough interest.

I built a tennis app. Would love feedback. If people find it useful, I'd be happy to share the weights and the codebase.

Location: Denver, CO

Willing to Relocate: No

Technologies:

    - AI/CV/NLP. Founder/CEO of Shotcount, a basketball/AI iOS app with 5,000+ downloads
    - Product Management. Founder/CEO of AR startup and raised $300,000
    - Full-stack web development. Most recently I built a next.js/express/GraphQL Shopify competitor, which helped the company raise $230,000

Résumé/CV:

    https://stev3.github.io/
    https://twitter.com/stevejenkinsyo
    http://github.com/stev3
    steven [at] stevenjenkins [dot] io

--

Hi HN! 10 years of product and engineering experience. I'm looking for technical product management opportunities in health/fin/urban -tech. I've been through several company and product cycles and know how to bring new products to market and scale them.

Thanks for all the hard work and congrats on your launch!

I will definitely check this service out for a side project I'm working on that combines basketball and AI (https://www.myshotcount.com/)

Thanks for sharing

It's a continuous take, but probably a couple were tossed before that one ;-)

Congrats on the launch, Andrew! I'm curious to learn more about your experience in mobile 3D reconstruction. Did you publish anything? I'm curious because I'm working in the 3D reconstruction space as well.

Thanks, man. I built it, showed some friends, and then learned there was a competitor, but that's cool. The world needs more than one and for some reason they don't offer anything to the billions of android users.

Based on my conversations with users, shot tracking is most used feature by far. There are a bunch of other services that help with dribbling [0] and drills [1].

[0] https://dribbleup.com/ [1] https://www.94feetofgame.com/app <- also a Steve Nash project

Apparently I'm bad at drawing penises.

Thanks! Trying to make it as simple as possible.

Since I can't play basketball with people, I built an app that helps me play basketball with myself. It uses object detection to identify me shooting and the ball going in or not and then creates a heatmap in real-time. Sort of like fitbit for basketball. I had to do some labelling myself, but it didn't take too long and it's working! I try to beat my shooting percentage from the day before. I put it on the Apple App Store to start and I'll build an android version next.

You can see a live demo here: https://www.myshotcount.com/

For people interested in identifying plants, the PlantCLEF dataset is a great start. Here's the link: https://www.imageclef.org/PlantCLEF2019

Implement a simple image classifier built with fast.ai and you can go quite far!

I believe it is to be used for research only.

Yes, exactly. :-)