Skip to main content
All events

Make your media and AI output type-safe with Pixeltable

You already know type safety from Pydantic. Pixeltable extends it to the images, video and LLM output you store, and guarantees that everything derived from them is made the same way for every file. We build a secondhand marketplace app with Pixeltable as its backend: sellers post a price and photos, Gemini writes the title, description and condition into a Pydantic model, and shoppers search or ask for "a like-new lamp under $100". Every step shows up in Logfire.

Pixeltable extends Pydantic's type safety to the images, video and LLM output you store, with every step visible in Logfire.

Overview

  • Live workshop
  • Online event
  • Free

Speaker

About this event

You already know type safety from Pydantic. Pixeltable extends it to the images, video and LLM output you store, and guarantees that everything derived from them is made the same way for every file. We build a secondhand marketplace app with Pixeltable as its backend: sellers post a price and photos, Gemini writes the title, description and condition into a Pydantic model, and shoppers search or ask for "a like-new lamp under $100". Every step shows up in Logfire.

Who it's for: Python developers who use Pydantic and build apps that store images, video, audio, documents or LLM output.

Read ahead: Alison Hill's Let your Pydantic AI agent search your media files walks through the marketplace app this session builds.

What you'll learn

  • Pydantic models as the shape of what Pixeltable stores, including LLM output
  • Every LLM call Pixeltable makes, on every row, in Logfire
  • What happens to a broken or mislabeled file, and how the refusal shows up in Logfire
  • Query results and API responses back as Pydantic models
  • A Pydantic AI agent searching Pixeltable tables, each query traced under its tool call

Save your spot

Register today and we'll send you the link to join, reminders, and the recording afterwards.

Spots available

You're on the list

Check your inbox for joining details and reminders before the session starts.

We couldn't save your registration

Please go back and try again in a moment.

Get started with Logfire for free

Sign up today and see your first trace in under 2 minutes

01 Install

$ uv add logfire

Resolved 7 packages in 128ms
Installed 7 packages in 89ms
+ logfire

02 Instrument

import logfire

logfire.configure()
logfire.info('Hello {name}', name='world')

Free up to 10m records per month No credit card needed