GSoC 2026 - Interest in Add support for generate_series in PostgreSQL

Mentorship Google Season of Code
1

Hi everyone,

I’m Aarushi Bahri, an MS Data Science student at Stony Brook University (graduating May 2027), and I’m interested in working on the Add support for generate_series in PostgreSQL project for GSoC 2026.

A bit about my background: I spent 2+ years at Bank of America as a software engineer building ETL pipelines processing millions of daily records across 50+ upstream systems, writing complex SQL queries, and optimizing pipeline delivery timelines by 40%. I work primarily in Python and have hands-on PostgreSQL experience both professionally and in personal projects. I’m currently doing ML research at Stony Brook on clinical datasets, where I manage processed data in PostgreSQL for modeling and analysis.

I’ve read through the Django contribution guidelines and the referenced issues and forum thread for this project:

Before I draft a full proposal, I wanted to introduce myself and ask a few questions:

  1. The project description mentions that generic composite field support may be a stepping stone toward generate_series — should the proposal treat these as two sequential milestones within the same project, or is composite field support considered out of scope for GSoC?
  2. Is there a preferred API design direction already discussed — for example, should generate_series live as a new ORM expression, a custom queryset method, or something else?
  3. Are there any existing PRs or experiments on this that I should review before writing my proposal?

I’m planning to make some contributions to Django before the submission deadline to get more familiar with the codebase. Any pointers on good first areas related to contrib.postgres or the ORM would be really appreciated.

Thanks for your time — looking forward to the discussion!

Aarushi

1 Like