https://prefect.io logo
Title
a

Alexander Kolev

01/09/2020, 9:44 PM
Hey guys, just found about Prefect and reading through the docs, it seems like a good match for our use case (main pain point when using airflow is scheduling latencies). I have 2 questions though: 1. Is there anyone already using this in production or is it still in active development and the only stable bits being hosted by the creating company (the Prefect Cloud)? 2. Can anyone confirm they've used Prefect for streaming-like data processing, where DAGs need to be executed on demand and more frequently (vs long running, cron-scheduled runs)
πŸ‘‹ 2
πŸ‘€ 1
t

Tyler Wanner

01/09/2020, 10:36 PM
Hi Alexander! We're happy to have caught your interest and welcome to the community! There is certainly a community of "prefect-in-production" users, both inside and outside of this Slack group. Of course we at Prefect are one, but we're a little biased πŸ˜‰, so you'll probably want to hear from them... but one partner in particular loves Prefect so much they wrote a blog post about it--https://www.symphonyrm.com/galaxy-brain-data-science-workflows-with-prefect-dask/
:upvote: 1
b

Braun Reyes

01/10/2020, 1:02 AM
We use Prefect in Production at Clearcover. We have not written about it yet but hope do so in Q1. Though I am not sure I would personally use Prefect for real time stream processing though...however less frequent consumers would probably be a good use case.
Our use case has to do with providing a self service data workflow tool for our analysts and data scientists
j

Joe Schmid

01/10/2020, 4:04 AM
Hi @Alexander Kolev, I wrote the blog post mentioned above and am CTO at a healthcare data science startup. We run Prefect in production and have had a very positive experience both with the product and the Prefect team’s responsiveness & support. Happy to answer any questions you have either publicly or privately β€” slack me here any time. (We have Flows for both data science pipelines & ETL.)
a

Alexander Kolev

01/10/2020, 6:36 AM
Thx all guys! I think the next logical step for us is to prototype and test it ourselves. As mentioned yesterday - docs look promising and the community is active, so all in all - very positive signs πŸ™‚
πŸ‘ 3