Skip to content

Popoto 1.0.0 Launch Announcements

Hacker News — Show HN

How to post

  1. Go to https://news.ycombinator.com/submit
  2. Use the title and URL below
  3. Post Tuesday–Thursday, 9–11am ET for best visibility
  4. Immediately add your top-level comment (below)
  5. Stay online for the first 2 hours to reply to comments — HN rewards fast engagement

Title

Show HN: Popoto – A Redis/Valkey ORM with Django-like syntax for Python

URL

https://github.com/tomcounsell/popoto

Your first comment (post immediately after submitting)

Hey HN, I'm Tom. I built Popoto because I wanted Django-model ergonomics for Redis without the overhead of Redis OM or hand-rolling SCAN/pipeline logic everywhere.

Popoto gives you a declarative model layer on top of Redis (or Valkey):

from popoto import Model, KeyField, SortedField, Q

class Restaurant(Model):
    name = KeyField()
    cuisine = Field()
    rating = SortedField(type=float)

Restaurant.create(name="Burger Palace", cuisine="American", rating=4.5)

# Chainable queries, Q objects, expression syntax
results = Restaurant.query.filter(
    Q(cuisine="Italian") | Q(cuisine="Japanese"),
    rating__gte=4.0
).order_by("-rating").limit(10)

What makes it different:

  • Django-like API — KeyField, SortedField, GeoField, Relationship, Q objects, get_or_create(), bulk_create()
  • Full async/await — native redis.asyncio, not asyncio.to_thread() wrappers
  • Atomic saves via internal Redis pipelines
  • Works with both Redis and Valkey out of the box
  • Geo search, time-series, pub/sub built in
  • Zero required dependencies beyond redis-py and msgpack — pandas is optional

We've been using it in production for streaming financial data and sensor networks. 1.0.0 is the first stable release after two betas and significant index integrity hardening.

Docs: https://popoto.io PyPI: pip install popoto

Happy to answer any questions about the internals or how it compares to Redis OM.

Product Hunt

How to post

  1. Go to https://www.producthunt.com/posts/new
  2. You need a maker account (sign up or log in)
  3. Launch at 12:01am PT on a Tuesday, Wednesday, or Thursday
  4. Fill in the fields below
  5. Have 5–10 supporters ready to upvote and comment in the first hour
  6. Engage with every comment throughout launch day

Product name

Popoto

Tagline (60 chars max)

Django-like ORM for Redis and Valkey in Python

Topics/tags

Developer Tools, Open Source, Python, Databases

Description

Popoto is a Python ORM that brings Django-model syntax to Redis and Valkey. Define your data models with fields, relationships, and indexes — Popoto handles serialization, querying, and persistence automatically.

Why Popoto?

Redis is incredibly fast, but using it as a primary data store means writing a lot of boilerplate — key naming conventions, serialization, index management, pipeline batching. Popoto eliminates all of that with a familiar, declarative API.

Key features:

🔍 Chainable queries with Q objects and expression syntax ⚡ Full async/await with native redis.asyncio 🌍 Geographic distance search with GeoField 📊 Time-series data with SortedField 🔗 Model relationships with lazy loading 📦 Bulk create/update/delete via Redis pipelines 🔄 Works with both Redis and Valkey

Who is it for?

Python developers who use Redis for more than caching — real-time applications, streaming data pipelines, IoT sensor networks, or anywhere you want Redis speed with ORM convenience.

1.0.0 is production-stable after two beta releases and extensive index integrity hardening.

GitHub: https://github.com/tomcounsell/popoto
Docs:   https://popoto.io
PyPI:   https://pypi.org/project/popoto/

Maker comment (post on launch)

Hi Product Hunt! I'm Tom, the creator of Popoto.

I built this because every Redis project I worked on ended up with the same boilerplate — key naming, msgpack serialization, maintaining secondary indexes by hand. I wanted something that felt like Django models but ran on Redis.

Popoto started as an internal tool for processing streaming financial data, where we needed sub-millisecond reads and Django-like query ergonomics. After 3+ years in production and two beta releases, 1.0.0 is the first stable release.

I'd love to hear how you'd use it — or what's missing. Feedback welcome!

General tips

  • Lead with the problem, not the product. "Redis has no good ORM" resonates more than "here's my ORM."
  • A demo GIF is worth 1000 words. Consider recording a 30-second terminal session showing model definition → create → query → result. Tools: asciinema, vhs, or a simple screen recording.
  • Don't cross-post on the same day. Do HN first (higher risk, higher reward), then Product Hunt a few days later with learnings from HN feedback.
  • GitHub README is your landing page. Make sure it looks sharp before posting.