oussema2021/fintuned_v3_AiRecruter
oussema2021/fintuned_v3_AiRecruter is a fine-tuned Llama 3.1 8B Instruct model, developed by oussema2021, designed to simulate a professional technical recruiter named "Alex." This model specializes in conducting realistic, multi-turn, voice-style interviews grounded in specific candidate CVs and Job Descriptions. It excels at interview simulation, recruiter training, and HR tech prototyping by maintaining a natural conversational tone and asking probing follow-up questions.
Loading preview...
Model Overview
oussema2021/fintuned_v3_AiRecruter, also known as Alex-Tech-Recruiter-v3, is a specialized fine-tuned version of the Llama 3.1 8B Instruct model. Developed by oussema2021, this model is engineered to realistically simulate a professional technical recruiter named "Alex" for multi-turn interview roleplay.
Key Capabilities
- Realistic Interview Simulation: Acts as a practice partner for candidates, maintaining a natural conversational tone and asking probing follow-up questions across multiple turns.
- Recruiter Training: Demonstrates ideal probing techniques and active listening for junior recruiters, showcasing how to build rapport and structure technical conversations.
- HR Tech Prototyping: Serves as a core conversational engine for automated screening tools, chatbot-based interview platforms, and recruitment pipeline applications.
- Contextual Interviewing: Designed to conduct interviews grounded in provided candidate CVs and Job Descriptions.
Training Details
The model was fine-tuned using QLoRA (Quantized Low-Rank Adaptation) via the Unsloth framework, utilizing 4-bit quantization. It supports a maximum sequence length of 4096 tokens.
Limitations
Users should be aware of potential hallucination if CV/JD context is missing or ambiguous, length bias in extremely long conversations (>20 turns), and persona rigidity due to heavy fine-tuning on the "Alex" persona. This model is intended for simulation and prototyping, not autonomous hiring decisions, and requires human oversight in real-world hiring workflows.