- Full Traicie Selection Specialist Flow implemented

- Added Specialist basics for handling phases and automatically transferring data between state and output
- Added QR-code generation for Magic Links
This commit is contained in:
Josako
2025-06-23 11:46:56 +02:00
parent 5b2c04501c
commit 7b87880045
10 changed files with 272 additions and 33 deletions

View File

@@ -16,10 +16,10 @@ backstory: >
AI-driven sourcing. Youre more than a recruiter—youre a trusted advisor, a brand ambassador, and a connector of
people and purpose.
{custom_backstory}
full_model_name: "mistral.mistral-medium-latest"
full_model_name: "mistral.magistral-medium-latest"
temperature: 0.3
metadata:
author: "Josako"
date_added: "2025-06-18"
description: "HR BP Agent."
description: "Traicie Recruiter Agent"
changes: "Initial version"

View File

@@ -275,6 +275,7 @@ class DevConfig(Config):
# Define the nginx prefix used for the specific apps
EVEAI_APP_LOCATION_PREFIX = '/admin'
EVEAI_CHAT_LOCATION_PREFIX = '/chat'
CHAT_CLIENT_PREFIX = 'chat-client/chat/'
# file upload settings
# UPLOAD_FOLDER = '/app/tenant_files'

View File

@@ -0,0 +1,37 @@
version: "1.0.0"
name: "KO Criteria Interview Definition"
task_description: >
In context of a vacancy in your company {tenant_name}, you are provided with a set of competencies
(both description and title). The competencies are in between triple backquotes. The competencies provided should be
handled as knock-out criteria.
For each of the knock-out criteria, you need to define
- A short (1 sentence), closed-ended question (Yes / No) to ask the recruitment candidate. Use your experience to ask a question that
enables us to verify compliancy to the criterium.
- A set of 2 short answers (1 small sentence each) to that question (positive answer / negative answer), from the
candidates perspective.
The positive answer will result in a positive evaluation of the criterium, the negative answer in a negative evaluation
of the criterium. Try to avoid just using Yes / No as positive and negative answers.
Apply the following tone of voice in both questions and answers: {tone_of_voice}, i.e. {tone_of_voice_context}
Apply the following language level in both questions and answers: {language_level}, i.e. {language_level_context}
Use {language} as language for both questions and answers.
```{ko_criteria}```
{custom_description}
expected_output: >
For each of the ko criteria, you provide:
- the exact title as specified in the original language
- the question in {language}
- a positive answer, resulting in a positive evaluation of the criterium. In {language}.
- a negative answer, resulting in a negative evaluation of the criterium. In {language}.
{custom_expected_output}
metadata:
author: "Josako"
date_added: "2025-06-20"
description: "A Task to define interview Q&A from given KO Criteria"
changes: "Improvement to ensure closed-ended questions and short descriptions"