n8n Interview Prep Workflow
A research-stage n8n workflow that assembles company context and produces targeted interview-prep drills with manual source review.
What it is: A research workflow for turning scattered role and company context into reviewable interview-prep drills.
What I built: Explored the workflow structure, review checkpoints, and context-gathering logic for a narrower interview-prep system.
Current state: Research-stage work: the concept and architecture are ahead of the implementation maturity.
Why it matters: Collects role context from notes and selected public references into one prep flow.
Category: Research / Experiment
Status: Research
Visibility: Public
This is a research-stage portfolio entry, not a claim of fully shipped production implementation.
What this project is
A workflow concept for gathering company context and turning it into targeted interview drills instead of scattered prep notes.
Why I explored it
Interview prep is often a messy mix of notes, role pages, company research, and last-minute prompt writing. I wanted a narrower system that could collect the useful context and turn it into practice material without pretending the workflow should run fully unattended.
Constraints
- Human oversight needs to stay in the loop for high-stakes or ambiguous interpretation.
- The workflow should not hide where the underlying context came from.
- Output quality depends on keeping references reviewable and scoping the drills to the role.
Architecture direction
- Structured intake of role notes and selected references.
- n8n orchestration for context assembly and drill generation.
- Manual review gate before anything becomes final prep material.
Current state
This is still research work. The useful output so far is the workflow framing and review model, not a finished production tool.
Why it matters
The point is not to automate thinking away. It is to compress repetitive prep overhead into a cleaner drill-building workflow while keeping the reasoning chain inspectable.
Key decisions
- Keep source review manual so generated prompts stay anchored to real context.
- Model confidence labels explicitly instead of pretending all prompts are equally strong.
- Use orchestration only where it reduces prep friction without hiding important reasoning steps.
What I'd improve next
The next improvement would be a lightweight source-trace view so each drill can be tied back to the notes or references that produced it.