The author reports an unsuccessful Instacart backend interview process involving CodeSignal assessments and human-led coding. The first onsite segment included a 90-minute library-application exercise and two hour-long coding interviews. AI was available for the application task, but only to explain code and help localize bugs, not to provide direct fixes. The author says the required changes ultimately remained in backend code.
The live interviews covered variable evaluation from equations and purchase pricing with promotional offers. The author attributes the failure to the equation round, where input restrictions became clear only late in the discussion. Much of the allotted time was spent debating cases the interviewer later excluded, leaving no time for an arithmetic extension. The report criticizes that communication and states that the applicant did not advance to the later system-design portion.
Discussion
Loading comments…