Guides for dummies?

Hey all, I read a lot of documentation of Algolia and I think for someone experienced with search this is incredible. I am building a small project and I have very basic needs for now and the documentation is really not helpful in the sense that it does not explain how to do simple things through a lens of liek case studies.

Is there a forum or collection of guides for very simple use cases somewhere?

My case is simple:
I have a collection of documents like this one below.
I am trying to make it so that if a user enters “@Alexander” in search bar - Algolia will only look up for Alexander inside of metadata->dna->labels.

I have tried multiple things already but can’t get to understand how the syntax should work. I tried to add a query rule in the API;(

      "_id": "ff03ea69-45e2-46a8-9909-101393d83157",
      "rules": {
        "fees": {
          "clone": 1,
          "share": 1,
          "untag": 1
        "shared": true,
        "royalty": true,
        "unbreakable": false
      "metadata": {
        "dna": {
          "labels": [
          "authors": [
          "snippets": [
        "hex": "11fd12fdfb8aefa7d73b1d117e846189",
        "used": 0,
        "cloned": false,
        "created": 1541100903942,
        "reported": 0,
        "purchased": 0
      "tags": [
      "thumbnail": "blob:",
      "status": "created",
      "createdAt": "2018-11-01T19:35:23.663Z",
      "updatedAt": "2018-11-01T20:02:25.297Z",
      "objectID": "ff03ea69-45e2-46a8-9909-101393d83157"

Here is what I tried:


Still getting weird error and have no idea what is that about;/

Hi there,

Normally if you put a special characters in the separatorsToIndex field in the settings, you can use it in the query rules anchoring.
Unfortunately, there seems to be a bug on this feature, I’m currently talking with the developers of the feature.

In the meanwhile, you could probably do the same on your front-end, catch the @ and apply the query parameter automatically.


Thanks you, so basically the API configuration can be buggy and its best for now to use my frontend?

In this case it’s buggy yes :slight_smile:
So, yes using a workaround - for now - in the front-end is better