What's the most over-engineered thing currently sitting in your stack?

Cedar Marauder · 0 likes

we have a full event-sourcing architecture with kafka for what is functionally a CRUD app with maybe 50 daily active users. the original architect left the company 3 years ago and nobody has the courage to rip it out

Noah Meier · 0 likes

this is basically every 'we might need scale someday' decision made by someone who left before 'someday' ever arrived lol

Noah Campbell · 0 likes

microservices for 200 req/day is almost impressive in its commitment to the bit honestly, respect

Liam Wilson · 0 likes

we have 6 different auth providers integrated because every new hire wanted to 'modernize' it without removing the old one. nobody knows which ones are actually load bearing anymore

Isha Menon · 0 likes

a GraphQL layer in front of a REST API that only has 3 consumers, all internal, all written by the same team that could just call the REST api directly

Maya Williams · 0 likes

not me but a friend's company runs a multi-region active-active database setup for an internal tool used by 12 employees in one office. the AWS bill alone could pay for a part time employee

Ethan Brooks · 0 likes

the AWS bill vs part time employee comparison is sending me, so real

Daan Bakker · 0 likes

honestly sometimes its not even technical debt its just fear, nobody wants to be the one who breaks prod simplifying something that technically works

Siti Lestari · 0 likes

we have terraform modules generating terraform modules. i did not make this up. i wish i had

Luca Bianchi · 0 likes

the answer is always kubernetes for something that could be a single VM with a cron job, every single time, in every company i've worked at