API Support: support@radix.ai | Website: https://www.radix.ai
\nJob and profile matching using Artificial Intelligence.
\nTalent API is a system that allows retrieving jobs recommendations for a profile or profiles recommendations for a job.
\nThere are three types of entities involved:
job: a JSON document describing a job offer.profile: a JSON document describing a user profile.match: a JSON document describing a link between a job and a profile. Getting started Here is a description of aYou need a client_id and a client_secret in order to have access to the API. If you don't have credentials, please
\ncontact us at hello@radix.ai. These credentials will also come with the scope you are allowed to use.
Using these, you can authenticate at the /auth endpoint and receive an access_token. This access_token will be needed
\nfor the next requests.
The request should look like
\ncurl -X POST \\\n https://[server-url]/auth \\\n -H 'Content-Type: application/json' \\\n -d '{\n "client_id": "[client_id]", \\\n "client_secret": "[client_secret]", \\\n "scope":"tapi/ALL"\n }'\nThe response will be
\n{\n "access_token":"eyJjIs7iOiJ[...]vQZUUhw2fsg",\n "expires_in":3600,\n "token_type":"Bearer"\n}\nBefore being able to request recommendations, you need to provide data. The minimum is a profile and a job.
\nHere is a sample request of how to send us a job.
\nThe example uses the smallest set of fields necessary for the query to work, but you should fill as many of them as
\npossible.
The field client_data can be used to send fields that are specific to your corporation that cannot be mapped to the API
\nspecification. Please note that using client specific fields is discouraged if your fields are non-critical or can be
\ntranslated into predefined fields.
curl -X POST \\\n http://[server-url]/jobs \\\n -H 'Authorization: Bearer eyJjIs7iOiJ[...]vQZUUhw2fsg' \\\n -H 'Content-Type: application/json' \\\n -d '{\n "external_id" : "12345",\n "title": "A job title",\n "addresses": ["The job address"],\n "publication_datetime" : "2020-02-12 00:02:00",\n "expiration_datetime" : "2020-02-12 00:02:00",\n "may_be_recommended" : true,\n "may_receive_recommendations" : true\n}'\nYou will receive the following response, confirming that we received the job and processed it correctly:
\n{\n "message": "Job successfully received. Recommendations will be available once the job has been processed."\n}\nIn the same fashion, to send us a profile
\ncurl -X POST \\\n http://[server-url]/profiles \\\n -H 'Authorization: Bearer eyJjIs7iOiJ[...]vQZUUhw2fsg' \\\n -H 'Content-Type: application/json' \\\n -d '{\n "external_id" : "6789",\n "addresses": ["The addresses where the candidate lives."],\n "publication_datetime" : "2008-02-12 00:02:00",\n "expiration_datetime" : "2008-02-12 00:02:00",\n "may_be_recommended" : true,\n "may_receive_recommendations" : true\n}'\nAnd the response will be
\n{\n "message": "Profile successfully received. Recommendations will be available once the profile has been processed.",\n}\nNow that there is a job and a profile in the system, you can ask for a recommendation. For example, in order to get the
\nbest jobs for the profile you just uploaded, you can do the following request
curl -X GET \\\n http://[server-url]/profile/123456789/jobs \\\n -H 'Authorization: Bearer eyJjIs7iOiJ[...]vQZUUhw2fsg' \\\n -H 'Content-Type: application/json' \\\nThe response will contain a list of the best jobs for that profile, in the following form
\n[\n {\n "resource_id": 12345\n "score": 0.87\n "resource": {\n "external_id" : "12345",\n "title": "A job title",\n "addresses": ["The job address"],\n "publication_datetime" : "2020-02-12 00:02:00",\n "expiration_datetime" : "2020-02-12 00:02:00",\n "may_be_recommended" : true,\n "may_receive_recommendations" : true\n },\n "match_explanation": {\n }\n },\n ...\n]\nThese recommendations are based on the data that you included when uploading jobs and profiles to Talent API. To receive
\nhigh quality recommendations, make sure to fill as many fields as possible when uploading entities.
If you want the recommendation to improve based on your user behaviour, you need to send us match indicators. A match
\nindicator represents a link between a profile and a job. As an example, if the user with profile_id 6789 had a look at
\nthe job page with id 12345, you can send us the following match:
curl -X POST \\\n https://[server-url]/matches \\\n -H 'Authorization: Bearer eyJraWQiOi[...]Mou8GdpvnA'\n -H 'Content-Type: application/json' \n -d '{\n 'profile_id':'6789',\n 'job_id':'12345',\n 'match_actor':'CANDIDATE',\n 'match_datetime':'2020-03-14T21:37:38.414Z',\n 'match_type':'PAGEVIEW'\n }'\nThat way, we are aware of user interactions inside of your data.
\nTo have a look at what you currently sent us, call the status endpoint:
curl -X GET \\\n https://[server-url]/status \\\n -H 'Authorization: Bearer eyJraWQiOi[...]Mou8GdpvnA'\n -H 'Content-Type: application/json' \nAPI Support: support@radix.ai | Website: https://www.radix.ai
\nJob and profile matching using Artificial Intelligence.
\nTalent API is a system that allows retrieving jobs recommendations for a profile or profiles recommendations for a job.
\nThere are three types of entities involved:
job: a JSON document describing a job offer.profile: a JSON document describing a user profile.match: a JSON document describing a link between a job and a profile. Getting started Here is a description of aYou need a client_id and a client_secret in order to have access to the API. If you don't have credentials, please
\ncontact us at hello@radix.ai. These credentials will also come with the scope you are allowed to use.
Using these, you can authenticate at the /auth endpoint and receive an access_token. This access_token will be needed
\nfor the next requests.
The request should look like
\ncurl -X POST \\\n https://[server-url]/auth \\\n -H 'Content-Type: application/json' \\\n -d '{\n "client_id": "[client_id]", \\\n "client_secret": "[client_secret]", \\\n "scope":"tapi/ALL"\n }'\nThe response will be
\n{\n "access_token":"eyJjIs7iOiJ[...]vQZUUhw2fsg",\n "expires_in":3600,\n "token_type":"Bearer"\n}\nBefore being able to request recommendations, you need to provide data. The minimum is a profile and a job.
\nHere is a sample request of how to send us a job.
\nThe example uses the smallest set of fields necessary for the query to work, but you should fill as many of them as
\npossible.
The field client_data can be used to send fields that are specific to your corporation that cannot be mapped to the API
\nspecification. Please note that using client specific fields is discouraged if your fields are non-critical or can be
\ntranslated into predefined fields.
curl -X POST \\\n http://[server-url]/jobs \\\n -H 'Authorization: Bearer eyJjIs7iOiJ[...]vQZUUhw2fsg' \\\n -H 'Content-Type: application/json' \\\n -d '{\n "external_id" : "12345",\n "title": "A job title",\n "addresses": ["The job address"],\n "publication_datetime" : "2020-02-12 00:02:00",\n "expiration_datetime" : "2020-02-12 00:02:00",\n "may_be_recommended" : true,\n "may_receive_recommendations" : true\n}'\nYou will receive the following response, confirming that we received the job and processed it correctly:
\n{\n "message": "Job successfully received. Recommendations will be available once the job has been processed."\n}\nIn the same fashion, to send us a profile
\ncurl -X POST \\\n http://[server-url]/profiles \\\n -H 'Authorization: Bearer eyJjIs7iOiJ[...]vQZUUhw2fsg' \\\n -H 'Content-Type: application/json' \\\n -d '{\n "external_id" : "6789",\n "addresses": ["The addresses where the candidate lives."],\n "publication_datetime" : "2008-02-12 00:02:00",\n "expiration_datetime" : "2008-02-12 00:02:00",\n "may_be_recommended" : true,\n "may_receive_recommendations" : true\n}'\nAnd the response will be
\n{\n "message": "Profile successfully received. Recommendations will be available once the profile has been processed.",\n}\nNow that there is a job and a profile in the system, you can ask for a recommendation. For example, in order to get the
\nbest jobs for the profile you just uploaded, you can do the following request
curl -X GET \\\n http://[server-url]/profile/123456789/jobs \\\n -H 'Authorization: Bearer eyJjIs7iOiJ[...]vQZUUhw2fsg' \\\n -H 'Content-Type: application/json' \\\nThe response will contain a list of the best jobs for that profile, in the following form
\n[\n {\n "resource_id": 12345\n "score": 0.87\n "resource": {\n "external_id" : "12345",\n "title": "A job title",\n "addresses": ["The job address"],\n "publication_datetime" : "2020-02-12 00:02:00",\n "expiration_datetime" : "2020-02-12 00:02:00",\n "may_be_recommended" : true,\n "may_receive_recommendations" : true\n },\n "match_explanation": {\n }\n },\n ...\n]\nThese recommendations are based on the data that you included when uploading jobs and profiles to Talent API. To receive
\nhigh quality recommendations, make sure to fill as many fields as possible when uploading entities.
If you want the recommendation to improve based on your user behaviour, you need to send us match indicators. A match
\nindicator represents a link between a profile and a job. As an example, if the user with profile_id 6789 had a look at
\nthe job page with id 12345, you can send us the following match:
curl -X POST \\\n https://[server-url]/matches \\\n -H 'Authorization: Bearer eyJraWQiOi[...]Mou8GdpvnA'\n -H 'Content-Type: application/json' \n -d '{\n 'profile_id':'6789',\n 'job_id':'12345',\n 'match_actor':'CANDIDATE',\n 'match_datetime':'2020-03-14T21:37:38.414Z',\n 'match_type':'PAGEVIEW'\n }'\nThat way, we are aware of user interactions inside of your data.
\nTo have a look at what you currently sent us, call the status endpoint:
curl -X GET \\\n https://[server-url]/status \\\n -H 'Authorization: Bearer eyJraWQiOi[...]Mou8GdpvnA'\n -H 'Content-Type: application/json' \n