Data ops, data engineering, data development – oh my!
From new roles, teams, skills and processes, the hot topic on everyone’s mind is data ops. I started to notice the data ops emergence back in 2015 as companies began to look at agile development to spin up new data capabilities rapidly. Later, as data preparation entered the market, ETL developers were gravitating to these tools for quick data loading with transparency into newly formed analytic lakes. Step into today and running advanced analytics (or the sexier term today – machine learning) in real-time and there is a lot of talk about the challenge of moving and updating analytics models from lab environments into production settings.
There is no doubt that vendors like DataKitchen, Metis Machine and DataRobot are all messaging and offering workbenches and capabilities to support the data ops needs. And, there is certainly a lot of gray area in the data platform communities of Informatica and Talend or the data science workbenches like Cognitive Scale that position to help with the engineering and instrumentation of data pipelines and model deployments/refreshes. The goal being a one-click method to push models to production or ease the burden.
Further still, let’s not forget the DevOps workbenches and testing platforms…
But, is there a market for the data ops workbench? Or, are data engineers just more prevalent and equipped with the skills and ingenuity to more quickly help move models to production – using the tools, coding or both to just get it done?
Vendors, if you are out there – flood my comments section, inbox, briefing requests and LinkedIn to show what you got. You may just make it into my Q4 research and evaluation.
Forrester clients – what tools, workbenches and technologies do your data engineers and developers use to rapidly build and manage quality pipelines and push analytics models to production? Help me set the evaluation criteria for what you want. Shoot me an email!
Okay, I threw down the gauntlet. Who’s ready to leap?