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

Here are 2 examples from the Black Spatula project where we were able to detect major errors: - https://github.com/The-Black-Spatula-Project/black-spatula-p... - https://github.com/The-Black-Spatula-Project/black-spatula-p...

Some things to note : this didn't even require a complex multi-agent pipeline. A single shot prompting was able to detect these errors.


This black spatula case was pretty famous and was all over the internet. Is it possible that the AI is merely detecting something that was already in its training data?


This is the original work that detected the problem.


I found many sources on the internet that state that the error in the black spatula paper was discovered by John Schwarcz of McGill University while reviewing a research paper:

[1] https://www.expatgo.com/my/2024/12/15/an-error-in-arithmetic... [2] https://gizmodo.com/embarrassing-math-error-scared-the-pants...

Is John Schwarcz behind this Yesnoerror or the black spatula project?


Thats what happens when you delegate counting things to GPT-4o


Awesome re: supporting both TypeORM and Sequelize


Thanks. While we began with Express, we intend to add support for more ORMs, so that users can pick the one they want. We did come across Prisma and we are exploring supporting it.

For now, we support MySQL, PostgreSQL and SQLite, but are also planning to add support for Mongo. Do you have Mongo/NoSQL databases on your roadmap?

It would be great to connect on a zoom - I'm the co-founder of imagine, my email is anusheel@imagine.ai


Imagine looks super cool!

Prisma has Mongo support coming really soon, and we will be speaking more about it at https://www.mongodb.com/live in July.

Other NoSQL and relational databases to follow, but we have no timeline for that yet.


Yes we will generate configs and scripts that would allow a single command deployment to AWS Lambda and Google Cloud Functions. We already generate a docker container that is ready to be deployed to Cloud Run or Fargate.


Thanks. Which feature is the one you found to be the most useful? Anything you would like us to add that could help during the initial phases of starting projects?


Hi - would be great to share with the community why we chose Haskell


Didn't you post the same link earlier today? Why do you need to ask why it uses Haskell?

See this comment https://news.ycombinator.com/item?id=27219439 ... in particular it makes me think you are trying to game HN with multiple accounts. Please don't


Hey :) it’s not multiple accounts - this post is by one of the core contributors to the project who is a functional programming expert and I thought it would be helpful if he shared his perspective on why Haskell was chosen for this!

(I couldn’t see my earlier post for some reason on Show HN and I thought I accidentally deleted it, so I thought rather than me reposting it, it would be nice to have one of our contributors share his perspective).


Multiple accounts can be separate people coordinating efforts to push up stories on HN. It is especially frowned upon when you act as if you don't know each other in the discussion or as if you are not a project contributor.

This is a text book example of such behavior. The other post adds to the suspicious activity. Please refrain in the future


Imagine is a labor of love for my co-founder and me. I’ve been a software engineer for over a decade and we’ve overseen the build out of several (80+) software products over the years.

Each time we were acutely reminded of how much redundancy there exists while writing code, especially for commonly used components and features in software applications. Also, while we believe the current version of the no-code movement is awesome, we found it hard to scale these solutions for applications that required any reasonable level of custom logic.

So we built Imagine - a configurable, cross-framework platform for developers to easily generate high-quality code for common application components, which they can seamlessly integrate into their projects, without any lock-in - and allowing them to focus on writing custom business logic. Our goal has been to make the generated code clean, readable and following known software best practices (we love the 12—factor app philosophy!) - basically indistinguishable from what a skilled engineer would write.

To build the Imagine code generator, we initially considered building with regexes, but discarded that approach as it would have been highly error-prone. Given the high configurability we wanted to offer, we needed to be obsessed with correctness. To achieve this, we decided to use Haskell to build the code generator, as we were able to use Haskell’s built-in type system to represent rules to ensure correctness of the generated code. It also allowed us to enforce various constraints across an app’s specification using static checks.

In terms of features, our beta release allows you to define your starter settings, data models and relations via a UI and we instantly generate code for a dockerized project starter, ORMs, REST & GraphQL CRUD APIs and tests in both Django & Node.js.

We also have the following features on our roadmap: - Backend frameworks: - Node/Typescript - Backend features: - NoSQL databases (MongoDB, DynamoDB) - Support for migrations - Kubernetes deployment scripts - Logging and monitoring - Authentication (via multiple providers) - Payments integration (via multiple providers) - Frontend frameworks: - React (incl Redux, Hooks) - React native - Frontend features: - API calling - Views & components - Storybooks

We’d like to thank you in advance for trying our product out - we’re still early in development, and would love your feedback and suggestions as we push to make Imagine better everyday!


Great looking app and excellent implementation so far! Under the models section you should consider adding an "Import CSV" function. Each column in the spreadsheet would represent a model. The first row would be the Model's name. Subsequent rows would be field names. Teams that are ready to start coding ideally should have their data model already created in a spreadsheet or something that can export to CSV. Importing a CSV from a spreadsheet could help speed things along.


You should switch to First Republic Forex account. We saved ~4% by switching from Transferwise to FRB on our international transfers.


First Republic Bank


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

Search: