Hacker Newsnew | past | comments | ask | show | jobs | submit | amolgupta's commentslogin

For app developers considering tflite, a safer way would be to host the models on firebase and delete them when their job is done. It comes with other features like versioning for model updates, A/B tests, lower apk size etc. https://firebase.google.com/docs/ml/manage-hosted-models


That wouldn't help against the technique explained in the article, would it? Since the model makes it way into the device, it can be intercepted in a similar fashion.

I'm not quite sure I understand the firebase feature btw. From the docs, it's pretty much file storage with a dedicated API? I suppose you can use those models for inference in the cloud, but still, the storage API seems redundant.


I think the comment author means offering inference via Firebase, with the model never leaving the backend.

This works, just like ChatGPT works, but has the downside of 1. You have to pay the computing for every inference 2. Your users can't access it offline 3. Your users will have to use a lot of data from their mobile network operator. 4. Your inference will be slower

And since SeeingAI infers the model every second, your and your customers bill will be huge.


That's what I thought, but the link doesn't say anything about off-device inference, it's only about storing and retrieving the model. There's just one off-hand note about cloud inference.

In any case, yeah you can not download the model to the device at all, but then you have to deal with the other angle - making sure the endpoint isn't abused.

Maybe a hybrid approach would work - infer just part of the model (layers?) on the cloud, and then carry on the inference on the device? I'm not familiar with how AI models look like and work like exactly, but I feel like hiding even a tiny portion of the model would make it not usable in practice


Your second note is very interesting, having looked at the model myself this is very plausible.

For models which use a lot of input nodes, a lot of "hidden layers" and in the end just perform a softmax this may get infeasible because of the amount of data you would have to transfer.

You may have inspired a second article :)


In addition to the sibling comment this would require repeatedly re-downloading models when you want to use them, which sucks.


I pass a kotlin data class and ask chatGPT to return json which can be parsed by that class. Reduces errors with date-time parsing and other formatting issues and takes up lesser tokens than the approach in the article.


I have a habit of making short bullet points notes for everything work and personal, but for my eyes only. With GPT, I can convert them into things like:

- PR descriptions. ex, paste bullet points about the change and it converts it into something to help the reviewer.

- Plant UML diagrams of ideas. At times they are just a starting point template and I build upon them. I can paste these into technical docs or PRs or presentations later.

- Peer feedback: The raw bullet points can be converted into nice-to-read feedback which is not too direct or offensive or vague. Can iterate over it to tone it down or make a point stronger.

Other programming-related use cases - Test cases for code

- Converting android xml layouts to compose worked well

- A lot of Django code

- Identify performance issues or bugs in code (these tasks make me realize the amount of repetition there is in programming)

Others:

- Book recommendations on topics

- Rewording emails/slack messages


I have used it to build travel itineraries and was tempted to write a travel app around that. Until I realized that some of the hotels and places it recommends do not actually exist or have existed in the past. It overconfidently also publishes broken booking links to these fake hotels. I am hoping that with chatGPT plugins, it would get better.


http://wise.com | Multiple positions | Austin, Budapest, London, NY, Tallinn, Singapore | Visa sponsored | Full time

Wise (formerly TransferWise) is on a mission to bring transparency to finance, for people without borders. We charge as little as possible, and we always show you upfront. No hidden fees. No bad exchange rates. No surprises.

We are hiring for numerous positions to help us achieve our mission including back-end and mobile engineers, product managers, data scientists.

Check out the roles and apply here -> https://grnh.se/5fb85f581us


http://wise.com | Multiple positions | Budapest, London, NY, Tallinn, Singapore, (Visa sponsored) | Full time

Wise (formerly TransferWise) is on a mission to bring transparency to finance, for people without borders. We charge as little as possible, and we always show you upfront. No hidden fees. No bad exchange rates. No surprises.

We are hiring for numerous positions to help us achieve our mission including back-end and mobile engineers, product managers, data scientists.

Check out the roles and apply here -> https://grnh.se/5fb85f581us


Just for info, Wise doesn't sponsors visa for graduate positions


http://wise.com | Multiple positions | Budapest, London, NY, Tallinn, Singapore, (Visa sponsored) | Full time

Wise (formerly TransferWise) is on a mission to bring transparency to finance, for people without borders. We charge as little as possible, and we always show you upfront. No hidden fees. No bad exchange rates. No surprises.

After going public last year, we are hiring for numerous positions to help us achieve our mission including back-end and mobile engineers, product managers, data scientists.

Check out the roles and apply here -> https://grnh.se/5fb85f581us


http://wise.com | Multiple positions | Budapest, London, NY, Tallinn, Singapore, (Visa sponsored) | Full time

Wise formerly TransferWise is on a mission to bring transparency to finance, for people without borders. We charge as little as possible, and we always show you upfront. No hidden fees. No bad exchange rates. No surprises.

After going public last year, we are hiring for numerous positions to help us achieve our mission including backend and mobile engineers, product managers, data scientists.

Check out the roles and apply here -> https://grnh.se/5fb85f581us


http://wise.com | Multiple positions | Budapest, London, NY, Tallinn, Singapore, (Visa sponsored) | Full time

Wise (formerly TransferWise) is on a mission to bring transparency to finance, for people without borders. We charge as little as possible, and we always show you upfront. No hidden fees. No bad exchange rates. No surprises.

After going public last year, we are hiring for numerous positions to help us achieve our mission, including backend and mobile engineers, product managers, data scientists.

Check out the roles and apply here -> https://grnh.se/5fb85f581us


http://wise.com | Multiple positions | Budapest, London, NY, Tallinn, Singapore, (Visa sponsored) | Full time

Wise formerly TransferWise is on a mission to bring transparency to finance, for people without borders. We charge as little as possible, and we always show you upfront. No hidden fees. No bad exchange rates. No surprises.

After going public last year, we are hiring for numerous positions to help us achieve our mission including backend and mobile engineers, product managers, data scientists.

Check out the roles and apply here -> https://grnh.se/5fb85f581us


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: