@Kyle Rasmussen can you give some idea about what you made so far?

I've created fairly robust source models (modeled against a BigQuery destination): • Salesforce (common objects like Accounts/Contacts/Opportunities/Individuals/Campaigns/Users/etc. plus a bunch of things specific to the Nonprofit Success Pack/NPSP) • Mailchimp (including incremental insert-overwrite models for event-stream data) • HubSpot (mostly the Marketing tables, including incremental insert-overwrite models for event-stream data) The final marts are pretty use-case-specific for us, but our intermediates would probably be pretty workable as source models with some minor edits. And as I mentioned in the other channel, these will allow for schema evolution (fields are only hard-coded when they have to be transformed; everything else operates on reading fields from adapter object or using

SELECT * EXCEPT(...)

) Our team has developed some others, but without being able to handle field additions . . . so we'll likely be re-modeling those soon. We also have several custom connectors for nonprofit-specific software and CRMs (which we haven't yet submitted)

Hi everyone, I’m still figuring out what the plan will be. But if you have already done something, share it here so we can discuss paths we can follow.

What I’m doing is trying to leverage the existing models trying to make Airbyte ERD output. This way it would be possible to use Fivetran’s and have other analytics models built on top. Definitely your expertise about some sources will be really valuable to design this project.

I have similarly worked on sources for BigQuery destination. I built them using the Fivetran ERD Diagrams for their Data Models and building out similar models using Airbyte entities. • Stripe: Robustly taking in the financial entities, still need to model the subscription entities, which covers the Stripe Connect side of the business • Recurly: modeling both the subscription and financial entities

Anything I can do to help here? Would love to contribute, specifically for "shopify" and "bigcommerce" 🙂!

Hello everyone! Sorry the lack of updates 🎉 This week, we’ll start the

Airbyte dbt-package

. This event will differ from others as it be built outside the Airbyte repository. For the event specifically we’re going to work creating models (transformations using dbt) to make Airbyte output compatible with Fivetran analytical models. The scope will be limited to the 25 sources Airbyte has in the catalog and have existing models in Fivetran. At the moment I’m finishing to setup the repository it is going to receive the contributions 🙂 as integration tests is an important part to maintain open source projects. We decided to start doing this “converter/migration” as it can leverage a lot of use cases with Airbyte connectors and in the future we have plans to receive additional models directly to the Airbyte dbt package itself.

Folks we decided to postpone the event in one week. I’m going to start posting more information during the week octavia muscle !

I know many people are eager to begin! Tomorrow is the day 🎉 I’m going to finish small details in the new repository we’re going to use during the Hackathon. Later today, I will publish more details about the rules, the steps to complete the task, and how the event will proceed. • ✅ Integration tests are running against BigQuery, Snowflake, and Postgres (sorry Redshift and other data warehouses for now). • ✅ I created a tutorial explaining the steps I took to get Zendesk Support working. I hope this will be helpful during the event. • ✅ Getting really excited for the Hackathon! 👋

THE HACKATHON HAS STARTED! You can access the repository and comment on one of the open issues to be assigned. This is a new repo so can have some missing details please report to me so I can make the improvements. There are 17 tasks for this Hackathon. The goal is to create transformations using dbt to make Airbyte compatible with Fivetran (if you want to propose new models feel free to create a Github issue to start a discussion). For now the event will be from Aug 8 to Aug 31 (as we approach the deadline, we can discuss extending it by a week or two). 1. First, go to the tutorial and quickly check if the task appeals to you. 2. Next, go to the repository and claim an issue. Once assigned, you can start with Step 1: checking if all tables and columns exist. FAQ • Is previous knowledge of using the source necessary to create the models? ◦ Not really, but for sources using reports like Google or Facebook, I’d recommend having prior experience with the data. • Can I have 2 or more issues assigned to me at the same time? ◦ No, you can only work in a single issue each time. You can only request a new task after getting the first merged. Repository: https://github.com/airbytehq/airbyte-dbt-models Tutorial: https://github.com/airbytehq/airbyte-dbt-models/discussions/43 I’m going to be here helping anyone with questions and trying to unblock you when you got stuck. I ask you to not DM but keep discussions in public. For running tests I have a Postgres instance with all sources sample data for that situation you can DM me to get the credentials 🔑 . Have initial questions please let’s continue in the thread🧵 Let’s have some fun and let’s hack! airbyte rocket

10 minutes and we got our first claim @tope folorunso 🚗 🤯 !

GREAT START!!! First day and 16 of 18 tasks were claimed, this is huge! Only HubSpot and TikTok Marketing are waiting for someone to pick them up. Here is a good example of what is going to happen during Step 2 and 3: https://github.com/airbytehq/airbyte-dbt-models/pull/46 (Thanks @Balasubramanian T K for being so fast!) If you need access to sample data please DM me so can share a Postgres instance you can query and run integration tests 👋 See ya!

Hi coming late for this @[DEPRECATED] Marcos Marx more issues will be added?

This time won’t be possible Felix 😞

I’m catching up on all issue comments and pull requests opened over the weekend 🕒 octavia thanks ! If you feel stuck or blocked please use this thread 🧵 so folks here can help you got unblocked!

The first contribution has been merged. 🎉 Congratulations to @tope folorunso for creating the Amplitude dbt package!

i noticed all the dbt packages are in one repo - is the plan to break them up after the hackathon or keep it on a monorepo?

Folks I didn’t have time today to catch up with all pending reviews. This is my top priority for tomorrow! Sorry the delay keanu thanks !

What is the difference between the fivetran_<connector> and the fivetran_<connector>_source ?

Hello everyone 👋 I recently finished a project that consumed some time, and I will return to all pending reviews of your contributions as soon as possible. We’re going to extend the deadline to next week to all contributions 🎉 . If you’re blocked or need any guidance doing the contribution feel free to schedule some time with me. octavia wave

WOOOOOOOHOOOO 🎉 🏁 !!! The hackathon has ended. There are 3 ongoing implementations I hope to finish this week. Thank you to everyone who has participated and collaborated to start this project. Yesterday we had zero, and today we’re heroes 🦸 with 16 dbt packages for Airbyte connectors. I will send a message to everyone whose contributions were merged to initiate the payment process 💰 !!!

Can you send the repo for this? I am integrating dbt for a new client.

Hi Team, I am trying integrating Airbyte with the DBT CORE but it is not giving me any option to do so. Could you please help me with that? I am using Airbyte open source community version installed through abctl.