Automation library · Hiring

Screen candidates and send take-home assignments

Started by an event in a connected app, and works across Gmail, Google Sheets and in-memory vectors. It puts Claude and OpenAI behind a retrieval step, so answers are drawn from documents you supply rather than from the model alone. 17 steps, 1 branch: a working afternoon to understand properly.

MediumCalls a model: Claude, OpenAIRetrieval

17 stepsMedium: a working afternoon
1Branch: places the flow splits
0Loops or waits
3Services it connects to

The flow

Starts it 1Model 8Logic 3Service 5

The steps in the order they run, from what starts it.

  1. 1Run manuallyManual triggerStarts it
  2. 2Rubric document loaderDocument Default Data LoaderModel
  3. 3EmbeddingsEmbeddings Open AiModel
  4. 4Anthropic Chat ModelLm Chat AnthropicModel
  5. 5Structured score parserOutput Parser StructuredModel
  6. 6Set config + rubricSetLogic
  7. 7Load rubric into vector storeVector Store In MemoryModel
  8. 8Rubric retriever (tool)Vector Store In MemoryModel
  9. 9Score candidate vs rubricAgentModel
  10. 10Fix score formatOutput Parser AutofixingModel
  11. 11Get candidates from Google SheetGoogle SheetsService
  12. 12Flatten scores for SheetCodeLogic
  13. 13Write scores back to SheetGoogle SheetsService
  14. 14Get approved shortlistGoogle SheetsService
  15. 15SwitchSwitchLogic
  16. 16Email the take-home assignmentGmailService
  17. 17Human evaluationGmailService

How it's built

Starts on
An event in a connected app
Steps
17, with 1 branch
Size
Medium: a working afternoon
Connects to
Gmail, Google Sheets, in-memory vectors
Model
Claude, OpenAI
Retrieval
Answers are drawn from documents you supply, not from the model alone

Questions

What does it need to run?

An n8n account, plus your own Gmail, Google Sheets and in-memory vectors accounts, and a key for Claude and OpenAI. You connect your own accounts after importing it; none are included.

What starts it?

It starts on an event in a connected app.

Does it use AI?

Yes. It calls Claude and OpenAI. Its answers draw on documents you give it, not on the model alone. Model calls are billed by the provider, not by n8n.

How long does it take to set up?

Plan an afternoon. 17 steps is enough that you should read each one before running it.

How do I import it?

Copy the workflow, open a new workflow in n8n and paste. The steps appear on the canvas, ready for your own accounts.

Before you run it

Nothing here has been tested by us. It is somebody's working automation, read and described from its own steps. Treat the connections it wants as a list of accounts it will touch, and read the 17 steps before you point it at anything that matters.