The easiest way to create data pipelines in Python

Close, minimize, and expand buttons on top-left hand corner of windows on Mac OSX
1
"""
2
Prism project configuration
3
"""
4
5
# Imports
6
from prism.admin import generate_run_id, generate_run_slug
7
import logging
8
from pathlib import Path
9
10
# Project metadata
10
NAME = "My first project!"
10
AUTHOR = "John Snow"
10
VERSION = "1.0"
10
DESCRIPTION = "My very first Prism project."
10
10
# Admin
11
RUN_ID = generate_run_id()
11
...

Create, configure, and run your first DAG in less than 2 minutes.

Get Started
Github logo
Star us on Github
01
· FEATURES
fingerprint

Focus on your code, not your admin.

Eliminate boilerplate and spaghetti code. With Prism, you can focus on what matters — your pipeline's logic.
Python logo
Frictionless development
navigate_next
Prism's declarative programming model and flexible CLI enables users to create powerful workflows with ease. Start developing locally with a single command.
“Batteries included”
navigate_next
Focus on writing your core Python code. Let Prism take care of logging, retries, concurrency, and more.
Engineering best practices
navigate_next
Projects are structured as DAGs, where each task is a standalone file containing modular code.
Fun background, blue to green gradient
Close, minimize, and expand buttons on top-left hand corner of windows on Mac OSX
extract.py
1
import prism.task
2
‍
3
class Extract(prism.task.PrismTask):
4
    RETRIES: int = 2
5
6
    def run(self, tasks, hooks):
7
        # get some data here
8
        return data
Close, minimize, and expand buttons on top-left hand corner of windows on Mac OSX
transform.py
1
import prism.task
2
‍
3
class Transform(prism.task.PrismTask):
4
5
    def run(self, tasks, hooks):
8
        data = tasks.ref("extract")
6
        # do some transformations here
7
        return transformed_data
Fun background, blue to red gradient
Close, minimize, and expand buttons on top-left hand corner of windows on Mac OSX
extract.py
1
import prism.task
2
‍
3
class Extract(prism.task.PrismTask):
4
    RETRIES: int = 2
5
6
    def run(self, tasks, hooks):
7
        # get some data here
8
        return data
Close, minimize, and expand buttons on top-left hand corner of windows on Mac OSX
transform.py
1
import prism.task
2
‍
3
class Transform(prism.task.PrismTask):
4
5
    def run(self, tasks, hooks):
8
        data = tasks.ref("extract.py")
6
        # do some transformations here
7
        return transformed_data
Fun background, purple to blue gradient
Close, minimize, and expand buttons on top-left hand corner of windows on Mac OSX
extract.py
1
import prism.task
2
‍
3
class Extract(prism.task.PrismTask):
4
    RETRIES: int = 2
5
6
    def run(self, tasks, hooks):
7
        # get some data here
8
        return data
Close, minimize, and expand buttons on top-left hand corner of windows on Mac OSX
transform.py
1
import prism.task
2
‍
3
class Transform(prism.task.PrismTask):
4
5
    def run(self, tasks, hooks):
8
        data = tasks.ref("extract.py")
6
        # do some transformations here
7
        return transformed_data
Cool, lined background pattern with a tech-y feel
Snowflake logo
Google BigQuery logo
Redshift logo
PostgreSQL logo
Trino logo
dbt logo
Apache Spark Logo
Docker logo
PrestoDB logo
Dask logo
Coming soon!
02
· INTEGRATIONS
integration_instructions

Integrate with your favorite third party data tools.

No more dealing with buggy, confusing operators. Integrate with the tools you already use in your data stack using simple function calls.

02
· Built for developers, trusted by enterprises

Effortlessly deploy your projects with Prism Cloud

(coming soon!)

Fun background, blue to green gradient
Prism Cloud view showing a list of DAGs
app_shortcut
Unified view
Gain complete visibility and control of your projects in a single pane of glass. Run, manage, and execute your projects through our intuitive UI.
campaign
Branch deployments
Accelerate your release cycle with separate deployment environments for your staging and production branches.
dashboard_customize
Custom Agents
Create separate Docker images for each project, or run each project on a dedicated, ephemeral virtual machine.
tips_and_updates
Managed service

Offload your operational concerns. Focus on writing and testing your projects — leave the infrastructure to us.

We’re still building, but we can keep you updated! 👇

Interested in learning more?

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.