Pinned Loading

1. drf-stripe-subscription drf-stripe-subscription Public

   An out-of-box Django REST framework solution for payment and subscription management using Stripe.

   Python 120 19

2. django-ninja-simple-jwt django-ninja-simple-jwt Public

   Simple and elegant stateless JWT-based authentication built for django-ninja

   Python 31 6

3. eau-de-go eau-de-go Public template

   Go template for web

   Go 8 1

4. building-efficient-api building-efficient-api Public

   Comparing API performance: Django REST Framework, Django Ninja, FastAPI, and Go

   Python 65 2

5. Building Django project like a Java ... Building Django project like a Java developer: design pattern for complex web projects

   1

   ## Background

   2

   Django and Django REST Framework are designed around Active Records design pattern where each Record Object represents a “living” database record that can be interacted with where the changes as resulted of the interaction is reflected on the underlying database record automatically. 

   3

   This has allowed many of Django's libraries including Django REST Framework to access data and modify data from all parts of the application, and thus encourages vertically integrated features where behaviors that are defined by Models, such as using ModelSerializer and ModelViewset.

   4

   5

   ## Challenges

6. Effcient hybrid search on vector dat... Effcient hybrid search on vector database using inline filtering with ScaNN index

   1

   ## Background

   2

   Filtering on structured metadata as well as embedding vectors effciently has historically been a scaling challenge for applications involving RAG. AlloyDB released a new feature called [inline filtering](https://cloud.google.com/blog/products/databases/enhancing-alloydb-vector-search-with-inline-filtering-and-enterprise-observability) to achieve exactly that. This gist contains some of my learnings while experimenting with inline filtering using ScaNN index to achieve efficent and scalable hybrid search.

   3

   4

   ## Summary (TLDR)

   5

   - The recommended query utilizes a 2-stage hybrid search process, where first stage performs a search on embedding chunks using inline filtering with ScaNN index, and the second stage refines the result by selecting the highest score chunk for each document.