The modern data stack has come a long way.
And yet...
To do anything interesting with data, you have to immerse yourself in a specialized ecosystem of solutions: from storage (e.g. snowflake, s3), to streaming (e.g. kafka), to processing (eg. spark, python scripts), to ingest (eg. fivetran, airbyte), to modeling (eg. dbt), to orchestration (eg. airflow). You have to hire as many data engineers as you can manage. You have to create data models, tables, views, topics, jobs, scripts, connectors, APIs, and SDKs - and then string them all together to play nicely in production. And good luck when something changes upstream, or you have to hand off work across teams.
By comparison, developing web applications and services is a breeze.
Software developers don't manage fragmented components in isolation and cross their fingers, hoping everything plays nicely. They create holistic applications, with contracts between services. They don't manually configure in production and in the cloud, spinning plates to keep everything alive. They define their infrastructure as code, and deploy with a single command. They use frameworks that abstract away infrastructure and middleware complexity. They leverage decades of software best practices for speed, quality and collaboration - like version control, CI/CD, automated testing, change management, and DevOps.
Why should development on a data/analytics stack be any different? It shouldn't be!
It. Should. Not. Be.
Our mission at Fiveonefour is to bring incredible developer experiences to the data & analytics stack. We believe that we'll have accomplished this when data or analytics can be dropped from the title of the people using data platforms. When data engineers are just engineers. When data services are just another software service - written by software engineers, powered by microservices and frameworks, and managed with software development best practices. When every developer can deliver high-quality data products, derive insights for themselves and their stakeholders, and easily integrate predictive and generative AI. When becoming data-driven is second nature, and ROI on data investments is nearly guaranteed.
Building software is a delight *. It's time to bring some delight to the data & analytics stack.
—Tim, Nico, Alex, Chris, Dan, Carlos, Olivia, Jonathan, Arman, the Georges, Joj, Isa, Michael, LJ, Benoit and Mitchell.
*Except for two things of course: naming things, cache invalidation and off by one errors.