Engineering ETL pipeline for DORA metrics with Faros AI
Faros AI is an engineering operations platform that connects incident logs through to version control, providing engineer managers observability into their software development lifecycle and other DORA metrics.
Interview with Thomas GERBER
How is Faros AI using Hasura?
Faros uses Hasura as the API layer between all their services. Previously, they built and maintained their own GraphQL service layer between the different components. Looking to create an open-sourced version of their application, they looked to Hasura and quickly replaced their entire API layer with Hasura within 4 weeks.
The Faros application ingests data through services like Airbyte, then writes that data with Hasura into a Postgres database, where a Metabase instance ingests, slices, and displays data queried through Hasura. Then, an instance of n8n lets users define events triggered by or queried from Hasura.
Hasura allowed the team to move quickly and remove all of their boilerplate code, unifying the entire API interface across all services.
"It was purely a matter of doing the automation in the automation tool, there was nothing else we had to do in the GraphQL endpoint because Hasura had done all the heavy lifting."
Software Engineer at Faros
GraphQL API (UPSERTS)
The auto-generated GraphQL API allowed the team to focus on their data model as their core, differentiated business offering, and then, quickly scaffold an API that matched their needs.
In particular, using Upsert mutations allowed the team to uniquely address the challenges present in engineering observability data collection, creating a competitive advantage compared to alternative solutions.
The team had a number of legacy services that would only pipe content through a REST endpoint, but being able to restify some endpoints at Hasura allowed the team to seamlessly integrate with those services with zero operations overhead.
Visit Faros AI website