What I Learned After One Year of Building a Data Platform Fr...

What I Learned After One Year of Building a Data Platform From Scratch

RelatedHighlights

You need to be able to draw insight and to structure that information such that people can then act on it. What is doing that? Today, transformation tools like dbt are doing that, if you take the lens of the data team really owning everything end-to-end, but I think also applications that are able to plug into the data warehouse, consume this raw i... See more

Jan-Erik Asplund • Earl Lee, co-founder and CEO of HeadsUp, on the modern data stack value chain

One of the problems with owning your own data is that it ignores the question of what format that data is in. Data formats are sometimes considered carefully, but they can also be hammered out fast to get a minimum viable product out the door.

Byrne Hobart • The Promise and Paradox of Decentralization

Get the AI-First product in the hands of customers without embarking on a yearlong data-lake project by manually pulling only the necessary data. This may be somewhat costly on a per-project basis but better than never finishing the project at all.

Ash Fontana • The AI-First Company: How to Compete and Win with Artificial Intelligence

Traditional ETL solutions are still quite powerful when it comes to:

Common connectors with small-medium data volumes : we still have a lot of respect for companies like Fivetran, who have really nailed the user experience for the most common ETL use cases, like syncing Zendesk tickets or a production Postgres read replica into Snowflake. The only

Why you should move your ETL stack to Modal

Everything starts slower on the data side. Building a valuable data asset takes time. Building the supporting infrastructure to actually deliver that data takes time. Sales cycles take time.

Abraham Thomas • The Economics of Data Businesses

My top issues with the foundations of the "modern data stack":

Snowflake - you may only pay for what you use, but you pay through the nose for it. The cost is so high that in many cases it can exceed the cost of binders full of dbas.