heading into the Okta data engineering / platform loop in about 3 weeks. the role is on the data infrastructure side, specifically around their analytics pipeline. recruiter mentioned 4 rounds after the coding screen but wasn't specific.
would love to hear from anyone who's gone through the loop recently, especially for non-SWE roles or infrastructure-adjacent ones. interested in: what kinds of technical problems showed up (SQL? pipeline design? pure coding?) how much identity/product knowledge they actually test vs general DE fundamentals behavioral focus areas, if any
any data points welcome, even vague ones are helpful at this stage.
4 replies
infra_ines
not okta specifically but i went through a data infra loop at an identity-adjacent company last year. the system design was almost entirely about pipeline architecture: how do you handle schema evolution, what's your strategy for backfills, how do you make a pipeline observable. no identity-specific knowledge required, they just want infra fundamentals.
numbers_only
saw a handful of okta DE interview reports on levels.fyi. patterns i noticed: SQL window functions came up multiple times, one person reported a Spark optimization problem, and the behavioral questions were mostly around handling production incidents and stakeholder communication. nothing that screams 'know oauth.'
de_derek
the window functions thing is useful, i'll make sure those are sharp. production incident storytelling i can do in my sleep, that's just my life.
sdr_sky
worth knowing: okta is going through some internal restructuring after the identity products consolidation. a few folks i know said teams were shifting mid-hire and headcounts moved around. worth asking your recruiter explicitly if the req is backfill or new headcount and whether the team structure is stable. not meant to scare you, just useful context.