The process included two one-hour live coding interviews focused on algorithmic exercises with progressively more complex requirements.
I completed the initial parts of both exercises, identified appropriate algorithms and data structures, and explained how I would handle memory constraints, streaming, and concurrency edge cases. AI is not allowed at all on the first part.
The second interview was particularly frustrating. We spent a significant amount of time debugging channel synchronization, including discussing interviewer suggestions that were incompatible with the Go version provided or could introduce additional concurrency issues.
What surprised me most was the written feedback. Alongside comments about implementation speed and debugging, one of the criticisms from Staff Engineer was:
"We would have liked more descriptive variable and function names to improve code readability."
It's 2026. AI coding agents can already generate, refactor, test, and polish substantial amounts of code. Of course, readability, correctness, and debugging still matter. But treating variable naming in a time-constrained coding exercise as a meaningful signal of senior engineering ability feels disconnected from modern software development.
For a senior backend/platform role, I would have expected more emphasis on architecture, distributed systems, production experience, engineering tradeoffs, and the ability to validate and deliver reliable solutions using modern tools.
My impression was that the process valued flawless execution of small coding exercises more than the engineering judgment required to build and operate real systems.
The following metrics were computed from 2 interview experiences for the Algolia Senior Software Engineer role in France.
Algolia's interview process for their Senior Software Engineer roles in France is extremely selective, failing the vast majority of engineers.
Candidates reported having very negative feelings for Algolia's Senior Software Engineer interview process in France.